{ "de-conversation-01-base17-segment-timed-u05-c00": { "id": "de-conversation-01-base17-segment-timed-u05-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83181, "plain": "Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir.", "source_duration_s": 4.0, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 188160, "script_start_s": 0.0, "script_end_s": 3.92, "all_asr_last_word_s": 3.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5772302695598899, "source_start_sample": 0, "source_end_sample": 192000, "first_word_output_sample": 0, "last_word_output_sample": 188160, "samples": 195360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-timed-u05-c00.mp3" }, "de-conversation-03-base17-segment-timed-u05-c00": { "id": "de-conversation-03-base17-segment-timed-u05-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85181, "plain": "Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt.", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 11, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 203520, "script_start_s": 0.16, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.4236287262994722, "source_start_sample": 3840, "source_end_sample": 207360, "first_word_output_sample": 3840, "last_word_output_sample": 199680, "samples": 206880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-timed-u05-c00.mp3" }, "de-conversation-04-base17-segment-timed-u02-c00": { "id": "de-conversation-04-base17-segment-timed-u02-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86133, "plain": "Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 13, "coverage": 0.8666666666666667, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.13333333333333333, "source_identity_reward": 0.5421150759976184, "source_start_sample": 0, "source_end_sample": 222720, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-timed-u02-c00.mp3" }, "de-conversation-02-base17-segment-free-u08-c00": { "id": "de-conversation-02-base17-segment-free-u08-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 11.84, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 11, "coverage": 0.8461538461538461, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 322560, "script_start_s": 0.16, "script_end_s": 6.72, "all_asr_last_word_s": 11.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.7692307692307693, "source_identity_reward": 0.2969812008538214, "source_start_sample": 3840, "source_end_sample": 329760, "first_word_output_sample": 3840, "last_word_output_sample": 318720, "samples": 325920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 4.97, "added_silence_s": 0.0, "duration_s": 6.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-free-u08-c00.mp3" }, "de-conversation-02-base17-segment-timed-u05-c00": { "id": "de-conversation-02-base17-segment-timed-u05-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84181, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören.", "source_duration_s": 5.36, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 249600, "script_start_s": 0.16, "script_end_s": 5.2, "all_asr_last_word_s": 5.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.720753339031612, "source_start_sample": 3840, "source_end_sample": 256800, "first_word_output_sample": 3840, "last_word_output_sample": 245760, "samples": 252960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-timed-u05-c00.mp3" }, "de-conversation-04-base17-segment-free-u09-c00": { "id": "de-conversation-04-base17-segment-free-u09-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 480000, "script_start_s": 0.24, "script_end_s": 10.0, "all_asr_last_word_s": 13.200000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5, "source_identity_reward": 0.3987238724298403, "source_start_sample": 7680, "source_end_sample": 487200, "first_word_output_sample": 3840, "last_word_output_sample": 472320, "samples": 479520, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 19.85, "added_silence_s": 0.0, "duration_s": 9.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-free-u09-c00.mp3" }, "en-conversation-01-base17-segment-free-u10-c00": { "id": "en-conversation-01-base17-segment-free-u10-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 9.28, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 257280, "script_start_s": 0.0, "script_end_s": 5.36, "all_asr_last_word_s": 5.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4630793919906731, "source_start_sample": 0, "source_end_sample": 264480, "first_word_output_sample": 0, "last_word_output_sample": 257280, "samples": 264480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 3.77, "added_silence_s": 0.0, "duration_s": 5.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-free-u10-c00.mp3" }, "de-conversation-02-base17-segment-free-u01-c00": { "id": "de-conversation-02-base17-segment-free-u01-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84117, "plain": "Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet.", "source_duration_s": 8.48, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 345600, "script_start_s": 0.24, "script_end_s": 7.2, "all_asr_last_word_s": 7.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7375503772470697, "source_start_sample": 7680, "source_end_sample": 352800, "first_word_output_sample": 3840, "last_word_output_sample": 337920, "samples": 345120, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 1.13, "added_silence_s": 0.0, "duration_s": 7.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-free-u01-c00.mp3" }, "de-conversation-03-base17-segment-timed-u08-c00": { "id": "de-conversation-03-base17-segment-timed-u08-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7283597012443559, "source_start_sample": 0, "source_end_sample": 222720, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-timed-u08-c00.mp3" }, "de-conversation-01-base17-segment-free-u06-c00": { "id": "de-conversation-01-base17-segment-free-u06-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83197, "plain": "Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 314880, "script_start_s": 0.0, "script_end_s": 6.56, "all_asr_last_word_s": 18.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.38461538461538464, "source_identity_reward": 0.4689189507633795, "source_start_sample": 0, "source_end_sample": 322080, "first_word_output_sample": 0, "last_word_output_sample": 314880, "samples": 322080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.29, "added_silence_s": 0.0, "duration_s": 6.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-free-u06-c00.mp3" }, "en-conversation-03-base17-segment-timed-u00-c00": { "id": "en-conversation-03-base17-segment-timed-u00-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75101, "plain": "Here is the part that interests me: using a short recording to establish each voice.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 234240, "script_start_s": 0.0, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06666666666666667, "source_identity_reward": 0.3983139437344396, "source_start_sample": 0, "source_end_sample": 238080, "first_word_output_sample": 0, "last_word_output_sample": 234240, "samples": 241440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-timed-u00-c00.mp3" }, "de-conversation-04-base17-block-timed-u01-c00": { "id": "de-conversation-04-base17-block-timed-u01-c00", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 86617, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.12, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 63, "coverage": 0.9692307692307692, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1098240, "script_start_s": 0.16, "script_end_s": 22.88, "all_asr_last_word_s": 22.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03076923076923077, "source_identity_reward": 0.26082256436877604, "source_start_sample": 3840, "source_end_sample": 1105440, "first_word_output_sample": 3840, "last_word_output_sample": 1094400, "samples": 1101600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 22.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u01-c00.mp3" }, "de-conversation-04-base17-block-timed-u01-c01": { "id": "de-conversation-04-base17-block-timed-u01-c01", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 86618, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.12, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 64, "coverage": 0.9846153846153847, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1102080, "script_start_s": 0.24, "script_end_s": 22.96, "all_asr_last_word_s": 22.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03076923076923077, "source_identity_reward": 0.2781515291762822, "source_start_sample": 7680, "source_end_sample": 1109280, "first_word_output_sample": 3840, "last_word_output_sample": 1094400, "samples": 1101600, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u01-c01.mp3" }, "de-conversation-04-base17-block-timed-u01-c02": { "id": "de-conversation-04-base17-block-timed-u01-c02", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 86619, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.12, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 61, "coverage": 0.9384615384615385, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1102080, "script_start_s": 0.16, "script_end_s": 22.96, "all_asr_last_word_s": 22.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06153846153846154, "source_identity_reward": 0.27568734819710233, "source_start_sample": 3840, "source_end_sample": 1109280, "first_word_output_sample": 3840, "last_word_output_sample": 1098240, "samples": 1105440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u01-c02.mp3" }, "de-conversation-04-base17-block-timed-u01-c03": { "id": "de-conversation-04-base17-block-timed-u01-c03", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 86620, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.2, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 63, "coverage": 0.9692307692307692, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1105920, "script_start_s": 0.16, "script_end_s": 23.04, "all_asr_last_word_s": 23.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03076923076923077, "source_identity_reward": 0.3677743221359966, "source_start_sample": 3840, "source_end_sample": 1113120, "first_word_output_sample": 3840, "last_word_output_sample": 1102080, "samples": 1109280, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u01-c03.mp3" }, "de-conversation-04-base17-block-timed-u01-c04": { "id": "de-conversation-04-base17-block-timed-u01-c04", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 86621, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.2, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 65, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1102080, "script_start_s": 0.16, "script_end_s": 22.96, "all_asr_last_word_s": 22.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.21370909124915133, "source_start_sample": 3840, "source_end_sample": 1109280, "first_word_output_sample": 3840, "last_word_output_sample": 1098240, "samples": 1105440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 23.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u01-c04.mp3" }, "de-conversation-04-base17-block-timed-u01-c05": { "id": "de-conversation-04-base17-block-timed-u01-c05", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 86622, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.2, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 63, "coverage": 0.9692307692307692, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1102080, "script_start_s": 0.08, "script_end_s": 22.96, "all_asr_last_word_s": 22.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03076923076923077, "source_identity_reward": 0.25810133953056047, "source_start_sample": 0, "source_end_sample": 1109280, "first_word_output_sample": 3840, "last_word_output_sample": 1102080, "samples": 1109280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 23.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u01-c05.mp3" }, "de-conversation-04-base17-block-timed-u01-c06": { "id": "de-conversation-04-base17-block-timed-u01-c06", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 86623, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.2, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 65, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1102080, "script_start_s": 0.24, "script_end_s": 22.96, "all_asr_last_word_s": 22.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.23489464890078732, "source_start_sample": 7680, "source_end_sample": 1109280, "first_word_output_sample": 3840, "last_word_output_sample": 1094400, "samples": 1101600, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 22.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u01-c06.mp3" }, "de-conversation-04-base17-block-timed-u01-c07": { "id": "de-conversation-04-base17-block-timed-u01-c07", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 86624, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.2, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 65, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1105920, "script_start_s": 0.24, "script_end_s": 23.04, "all_asr_last_word_s": 23.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.30976418405080713, "source_start_sample": 7680, "source_end_sample": 1113120, "first_word_output_sample": 3840, "last_word_output_sample": 1098240, "samples": 1105440, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u01-c07.mp3" }, "de-conversation-05-base17-segment-timed-u00-c00": { "id": "de-conversation-05-base17-segment-timed-u00-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87101, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch.", "source_duration_s": 4.0, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 10, "coverage": 0.9090909090909091, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09090909090909091, "source_identity_reward": 0.6425806973061595 }, "en-conversation-05-base17-block-free-u02-c00": { "id": "en-conversation-05-base17-block-free-u02-c00", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 77633, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 983040, "script_start_s": 0.16, "script_end_s": 20.48, "all_asr_last_word_s": 31.159999999999997 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.43636363636363634, "source_identity_reward": 0.07656764292386131, "source_start_sample": 3840, "source_end_sample": 990240, "first_word_output_sample": 3840, "last_word_output_sample": 979200, "samples": 986400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 39.37, "added_silence_s": 0.0, "duration_s": 20.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u02-c00.mp3" }, "en-conversation-05-base17-block-free-u02-c01": { "id": "en-conversation-05-base17-block-free-u02-c01", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 77634, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 833280, "script_start_s": 0.0, "script_end_s": 17.36, "all_asr_last_word_s": 33.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.23636363636363636, "source_identity_reward": 0.11071487186730518, "source_start_sample": 0, "source_end_sample": 840480, "first_word_output_sample": 0, "last_word_output_sample": 833280, "samples": 840480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 42.49, "added_silence_s": 0.0, "duration_s": 17.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u02-c01.mp3" }, "en-conversation-05-base17-block-free-u02-c02": { "id": "en-conversation-05-base17-block-free-u02-c02", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 77635, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 49.2, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 877440, "script_start_s": 0.16, "script_end_s": 18.28, "all_asr_last_word_s": 45.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2727272727272727, "source_identity_reward": 0.09511875321266623, "source_start_sample": 3840, "source_end_sample": 884640, "first_word_output_sample": 3840, "last_word_output_sample": 873600, "samples": 880800, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 30.77, "added_silence_s": 0.0, "duration_s": 18.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u02-c02.mp3" }, "en-conversation-05-base17-block-free-u02-c03": { "id": "en-conversation-05-base17-block-free-u02-c03", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 77636, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 954240, "script_start_s": 0.24, "script_end_s": 19.88, "all_asr_last_word_s": 19.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.13191337685411283, "source_start_sample": 7680, "source_end_sample": 961440, "first_word_output_sample": 3840, "last_word_output_sample": 946560, "samples": 953760, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 39.97, "added_silence_s": 0.0, "duration_s": 19.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u02-c03.mp3" }, "en-conversation-05-base17-block-free-u02-c04": { "id": "en-conversation-05-base17-block-free-u02-c04", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 77637, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 38.88, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 844800, "script_start_s": 0.16, "script_end_s": 17.6, "all_asr_last_word_s": 38.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.41818181818181815, "source_identity_reward": 0.07526512411515449, "source_start_sample": 3840, "source_end_sample": 852000, "first_word_output_sample": 3840, "last_word_output_sample": 840960, "samples": 848160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 21.13, "added_silence_s": 0.0, "duration_s": 17.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u02-c04.mp3" }, "en-conversation-05-base17-block-free-u02-c05": { "id": "en-conversation-05-base17-block-free-u02-c05", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 77638, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 810240, "script_start_s": 0.16, "script_end_s": 16.88, "all_asr_last_word_s": 33.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2909090909090909, "source_identity_reward": 0.12408628680129592, "source_start_sample": 3840, "source_end_sample": 817440, "first_word_output_sample": 3840, "last_word_output_sample": 806400, "samples": 813600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 42.97, "added_silence_s": 0.0, "duration_s": 16.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u02-c05.mp3" }, "en-conversation-05-base17-block-free-u02-c06": { "id": "en-conversation-05-base17-block-free-u02-c06", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 77639, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 40, "coverage": 0.7272727272727273, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2727272727272727, "source_identity_reward": 0.05433039494348879 }, "en-conversation-05-base17-block-free-u02-c07": { "id": "en-conversation-05-base17-block-free-u02-c07", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 77640, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 862080, "script_start_s": 0.16, "script_end_s": 17.96, "all_asr_last_word_s": 17.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.15312994794372903, "source_start_sample": 3840, "source_end_sample": 869280, "first_word_output_sample": 3840, "last_word_output_sample": 858240, "samples": 865440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 41.89, "added_silence_s": 0.0, "duration_s": 18.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u02-c07.mp3" }, "en-conversation-04-base17-segment-timed-u08-c00": { "id": "en-conversation-04-base17-segment-timed-u08-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 218880, "script_start_s": 0.16, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.16988666818512002, "source_start_sample": 3840, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-timed-u08-c00.mp3" }, "en-conversation-02-base17-segment-timed-u01-c00": { "id": "en-conversation-02-base17-segment-timed-u01-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74117, "plain": "Please tell me the tiny studio has a very dramatic red recording light.", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 203520, "script_start_s": 0.16, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2348647513918401, "source_start_sample": 3840, "source_end_sample": 207360, "first_word_output_sample": 3840, "last_word_output_sample": 199680, "samples": 206880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-timed-u01-c00.mp3" }, "en-conversation-02-base17-segment-free-u01-c00": { "id": "en-conversation-02-base17-segment-free-u01-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74117, "plain": "Please tell me the tiny studio has a very dramatic red recording light.", "source_duration_s": 11.76, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 234240, "script_start_s": 0.16, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3233168691073277, "source_start_sample": 3840, "source_end_sample": 241440, "first_word_output_sample": 3840, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 6.73, "added_silence_s": 0.0, "duration_s": 4.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-free-u01-c00.mp3" }, "en-conversation-01-base17-segment-timed-u07-c00": { "id": "en-conversation-01-base17-segment-timed-u07-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73213, "plain": "All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 13, "coverage": 0.9285714285714286, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 218880, "script_start_s": 0.16, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.14285714285714285, "source_identity_reward": 0.14150885739124205, "source_start_sample": 3840, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-timed-u07-c00.mp3" }, "en-conversation-02-base17-segment-timed-u08-c00": { "id": "en-conversation-02-base17-segment-timed-u08-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 218880, "script_start_s": 0.16, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.10584115038641116, "source_start_sample": 3840, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-timed-u08-c00.mp3" }, "de-conversation-03-base17-segment-free-u09-c00": { "id": "de-conversation-03-base17-segment-free-u09-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 353280, "script_start_s": 0.24, "script_end_s": 7.36, "all_asr_last_word_s": 7.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4815377788398747, "source_start_sample": 7680, "source_end_sample": 360480, "first_word_output_sample": 3840, "last_word_output_sample": 345600, "samples": 352800, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 22.49, "added_silence_s": 0.0, "duration_s": 7.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-free-u09-c00.mp3" }, "en-conversation-04-base17-segment-free-u10-c00": { "id": "en-conversation-04-base17-segment-free-u10-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 11.76, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 241920, "script_start_s": 0.0, "script_end_s": 5.04, "all_asr_last_word_s": 5.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.36340140353238437, "source_start_sample": 0, "source_end_sample": 249120, "first_word_output_sample": 0, "last_word_output_sample": 241920, "samples": 249120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 6.57, "added_silence_s": 0.0, "duration_s": 5.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-free-u10-c00.mp3" }, "de-conversation-01-base17-segment-timed-u11-c00": { "id": "de-conversation-01-base17-segment-timed-u11-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 218880, "script_start_s": 0.16, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5190990149209348, "source_start_sample": 3840, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-timed-u11-c00.mp3" }, "en-conversation-05-base17-segment-free-u11-c00": { "id": "en-conversation-05-base17-segment-free-u11-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.6, "source_identity_reward": 0.1737245607760952 }, "en-conversation-02-base17-segment-timed-u07-c00": { "id": "en-conversation-02-base17-segment-timed-u07-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74213, "plain": "Now that is a demonstration I can enjoy. Keep the little red light on.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 218880, "script_start_s": 0.16, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3548351946456183, "source_start_sample": 3840, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-timed-u07-c00.mp3" }, "en-conversation-01-base17-segment-free-u01-c00": { "id": "en-conversation-01-base17-segment-free-u01-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73117, "plain": "You sound suspiciously excited. Did a speech model actually surprise you?", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 276480, "script_start_s": 0.16, "script_end_s": 5.76, "all_asr_last_word_s": 5.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.350336582471675, "source_start_sample": 3840, "source_end_sample": 283680, "first_word_output_sample": 3840, "last_word_output_sample": 272640, "samples": 279840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 3.77, "added_silence_s": 0.0, "duration_s": 5.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-free-u01-c00.mp3" }, "de-conversation-01-base17-segment-free-u01-c00": { "id": "de-conversation-01-base17-segment-free-u01-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83117, "plain": "Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht?", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 10, "matched_tokens": 10, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 299520, "script_start_s": 0.0, "script_end_s": 6.24, "all_asr_last_word_s": 6.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4025443775923169, "source_start_sample": 0, "source_end_sample": 306720, "first_word_output_sample": 0, "last_word_output_sample": 299520, "samples": 306720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.61, "added_silence_s": 0.0, "duration_s": 6.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-free-u01-c00.mp3" }, "en-conversation-01-base17-segment-timed-u08-c00": { "id": "en-conversation-01-base17-segment-timed-u08-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 218880, "script_start_s": 0.16, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.35080317875943823, "source_start_sample": 3840, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-timed-u08-c00.mp3" }, "de-conversation-05-base17-segment-timed-u08-c00": { "id": "de-conversation-05-base17-segment-timed-u08-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.782679213556233, "source_start_sample": 0, "source_end_sample": 222720, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-timed-u08-c00.mp3" }, "de-conversation-05-base17-block-timed-u03-c00": { "id": "de-conversation-05-base17-block-timed-u03-c00", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 87649, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.96, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 68, "coverage": 0.9714285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1186560, "script_start_s": 0.16, "script_end_s": 24.72, "all_asr_last_word_s": 24.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05714285714285714, "source_identity_reward": 0.26580406044681076, "source_start_sample": 3840, "source_end_sample": 1193760, "first_word_output_sample": 3840, "last_word_output_sample": 1182720, "samples": 1189920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 24.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u03-c00.mp3" }, "de-conversation-05-base17-block-timed-u03-c01": { "id": "de-conversation-05-base17-block-timed-u03-c01", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 87650, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.96, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1186560, "script_start_s": 0.08, "script_end_s": 24.72, "all_asr_last_word_s": 24.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u03-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04285714285714286, "source_identity_reward": 0.22236142570314696, "source_start_sample": 0, "source_end_sample": 1193760, "first_word_output_sample": 3840, "last_word_output_sample": 1186560, "samples": 1193760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 24.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u03-c01.mp3" }, "de-conversation-05-base17-block-timed-u03-c02": { "id": "de-conversation-05-base17-block-timed-u03-c02", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 87651, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.96, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 68, "coverage": 0.9714285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1190400, "script_start_s": 0.16, "script_end_s": 24.8, "all_asr_last_word_s": 24.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u03-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05714285714285714, "source_identity_reward": 0.2352383318435131, "source_start_sample": 3840, "source_end_sample": 1197600, "first_word_output_sample": 3840, "last_word_output_sample": 1186560, "samples": 1193760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 24.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u03-c02.mp3" }, "de-conversation-05-base17-block-timed-u03-c03": { "id": "de-conversation-05-base17-block-timed-u03-c03", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 87652, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.96, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 68, "coverage": 0.9714285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1186560, "script_start_s": 0.0, "script_end_s": 24.72, "all_asr_last_word_s": 24.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u03-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05714285714285714, "source_identity_reward": 0.20529243723463775, "source_start_sample": 0, "source_end_sample": 1193760, "first_word_output_sample": 0, "last_word_output_sample": 1186560, "samples": 1193760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 24.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u03-c03.mp3" }, "de-conversation-05-base17-block-timed-u03-c04": { "id": "de-conversation-05-base17-block-timed-u03-c04", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 87653, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.96, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 68, "coverage": 0.9714285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1186560, "script_start_s": 0.16, "script_end_s": 24.72, "all_asr_last_word_s": 24.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u03-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05714285714285714, "source_identity_reward": 0.2950216434834536, "source_start_sample": 3840, "source_end_sample": 1193760, "first_word_output_sample": 3840, "last_word_output_sample": 1182720, "samples": 1189920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 24.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u03-c04.mp3" }, "de-conversation-05-base17-block-timed-u03-c05": { "id": "de-conversation-05-base17-block-timed-u03-c05", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 87654, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.96, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 67, "coverage": 0.9571428571428572, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1186560, "script_start_s": 0.16, "script_end_s": 24.72, "all_asr_last_word_s": 24.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u03-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07142857142857142, "source_identity_reward": 0.23130450126005211, "source_start_sample": 3840, "source_end_sample": 1193760, "first_word_output_sample": 3840, "last_word_output_sample": 1182720, "samples": 1189920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 24.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u03-c05.mp3" }, "de-conversation-05-base17-block-timed-u03-c06": { "id": "de-conversation-05-base17-block-timed-u03-c06", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 87655, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.96, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 68, "coverage": 0.9714285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1182720, "script_start_s": 0.16, "script_end_s": 24.64, "all_asr_last_word_s": 24.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u03-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05714285714285714, "source_identity_reward": 0.2296286455028441, "source_start_sample": 3840, "source_end_sample": 1189920, "first_word_output_sample": 3840, "last_word_output_sample": 1178880, "samples": 1186080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.17, "added_silence_s": 0.0, "duration_s": 24.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u03-c06.mp3" }, "de-conversation-05-base17-block-timed-u03-c07": { "id": "de-conversation-05-base17-block-timed-u03-c07", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 87656, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.96, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1186560, "script_start_s": 0.16, "script_end_s": 24.72, "all_asr_last_word_s": 24.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u03-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04285714285714286, "source_identity_reward": 0.20875253068124935, "source_start_sample": 3840, "source_end_sample": 1193760, "first_word_output_sample": 3840, "last_word_output_sample": 1182720, "samples": 1189920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 24.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u03-c07.mp3" }, "en-conversation-05-base17-segment-timed-u15-c00": { "id": "en-conversation-05-base17-segment-timed-u15-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77341, "plain": "Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 6.32, "source_frame_cap": false, "alignment": { "expected_tokens": 19, "matched_tokens": 19, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 295680, "script_start_s": 0.16, "script_end_s": 6.16, "all_asr_last_word_s": 6.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u15-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2700857304772412, "source_start_sample": 3840, "source_end_sample": 302880, "first_word_output_sample": 3840, "last_word_output_sample": 291840, "samples": 299040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 6.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-timed-u15-c00.mp3" }, "en-conversation-04-base17-segment-free-u00-c00": { "id": "en-conversation-04-base17-segment-free-u00-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76101, "plain": "I can write a requested length next to an individual sentence in the script.", "source_duration_s": 9.84, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 464640, "script_start_s": 0.16, "script_end_s": 9.68, "all_asr_last_word_s": 9.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3825431184284418, "source_start_sample": 3840, "source_end_sample": 471840, "first_word_output_sample": 3840, "last_word_output_sample": 460800, "samples": 468000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-free-u00-c00.mp3" }, "en-conversation-01-base17-segment-free-u02-c00": { "id": "en-conversation-01-base17-segment-free-u02-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73133, "plain": "It can read a script with different speakers and sentence-level acting directions.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.45370791129424026, "source_start_sample": 0, "source_end_sample": 226080, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 25.29, "added_silence_s": 0.0, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-free-u02-c00.mp3" }, "de-conversation-02-base17-segment-timed-u04-c00": { "id": "de-conversation-02-base17-segment-timed-u04-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84165, "plain": "Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 4.8, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 11, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.5535507958709985 }, "de-conversation-05-base17-segment-free-u15-c00": { "id": "de-conversation-05-base17-segment-free-u15-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87341, "plain": "Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 19, "matched_tokens": 19, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 311040, "script_start_s": 0.0, "script_end_s": 6.48, "all_asr_last_word_s": 6.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u15-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5187537526171815, "source_start_sample": 0, "source_end_sample": 318240, "first_word_output_sample": 0, "last_word_output_sample": 311040, "samples": 318240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.37, "added_silence_s": 0.0, "duration_s": 6.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u15-c00.mp3" }, "de-conversation-05-base17-segment-timed-u01-c00": { "id": "de-conversation-05-base17-segment-timed-u01-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87117, "plain": "Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus?", "source_duration_s": 4.48, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 10, "coverage": 0.8333333333333334, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.16666666666666666, "source_identity_reward": 0.5150041870382542 }, "en-conversation-01-base17-segment-free-u08-c00": { "id": "en-conversation-01-base17-segment-free-u08-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 9.04, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 268800, "script_start_s": 0.0, "script_end_s": 5.6, "all_asr_last_word_s": 5.6000000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.33661229436605017, "source_start_sample": 0, "source_end_sample": 276000, "first_word_output_sample": 0, "last_word_output_sample": 268800, "samples": 276000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 3.29, "added_silence_s": 0.0, "duration_s": 5.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-free-u08-c00.mp3" }, "en-conversation-01-base17-block-timed-u00-c00": { "id": "en-conversation-01-base17-block-timed-u00-c00", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 73601, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.64, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 791040, "script_start_s": 0.24, "script_end_s": 16.48, "all_asr_last_word_s": 16.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.23418061726396264, "source_start_sample": 7680, "source_end_sample": 798240, "first_word_output_sample": 3840, "last_word_output_sample": 783360, "samples": 790560, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u00-c00.mp3" }, "en-conversation-01-base17-block-timed-u00-c01": { "id": "en-conversation-01-base17-block-timed-u00-c01", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 73602, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.64, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 791040, "script_start_s": 0.0, "script_end_s": 16.48, "all_asr_last_word_s": 16.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.10963144168693359, "source_start_sample": 0, "source_end_sample": 798240, "first_word_output_sample": 0, "last_word_output_sample": 791040, "samples": 798240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u00-c01.mp3" }, "en-conversation-01-base17-block-timed-u00-c02": { "id": "en-conversation-01-base17-block-timed-u00-c02", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 73603, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.64, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 47, "coverage": 0.9038461538461539, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 794880, "script_start_s": 0.0, "script_end_s": 16.56, "all_asr_last_word_s": 16.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09615384615384616, "source_identity_reward": 0.18776058131048787, "source_start_sample": 0, "source_end_sample": 798720, "first_word_output_sample": 0, "last_word_output_sample": 794880, "samples": 802080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u00-c02.mp3" }, "en-conversation-01-base17-block-timed-u00-c03": { "id": "en-conversation-01-base17-block-timed-u00-c03", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 73604, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.64, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 791040, "script_start_s": 0.0, "script_end_s": 16.48, "all_asr_last_word_s": 16.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.1413238352353918, "source_start_sample": 0, "source_end_sample": 798240, "first_word_output_sample": 0, "last_word_output_sample": 791040, "samples": 798240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u00-c03.mp3" }, "en-conversation-01-base17-block-timed-u00-c04": { "id": "en-conversation-01-base17-block-timed-u00-c04", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 73605, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.64, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 48, "coverage": 0.9230769230769231, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 791040, "script_start_s": 0.24, "script_end_s": 16.48, "all_asr_last_word_s": 16.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07692307692307693, "source_identity_reward": 0.1347682554688405, "source_start_sample": 7680, "source_end_sample": 798240, "first_word_output_sample": 3840, "last_word_output_sample": 783360, "samples": 790560, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u00-c04.mp3" }, "en-conversation-01-base17-block-timed-u00-c05": { "id": "en-conversation-01-base17-block-timed-u00-c05", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 73606, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.64, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 791040, "script_start_s": 0.16, "script_end_s": 16.48, "all_asr_last_word_s": 16.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.1383035462733597, "source_start_sample": 3840, "source_end_sample": 798240, "first_word_output_sample": 3840, "last_word_output_sample": 787200, "samples": 794400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u00-c05.mp3" }, "en-conversation-01-base17-block-timed-u00-c06": { "id": "en-conversation-01-base17-block-timed-u00-c06", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 73607, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.64, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 49, "coverage": 0.9423076923076923, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09615384615384616, "source_identity_reward": 0.15918641045677884 }, "en-conversation-01-base17-block-timed-u00-c07": { "id": "en-conversation-01-base17-block-timed-u00-c07", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 73608, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.64, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 791040, "script_start_s": 0.16, "script_end_s": 16.48, "all_asr_last_word_s": 16.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.1603760336138189, "source_start_sample": 3840, "source_end_sample": 798240, "first_word_output_sample": 3840, "last_word_output_sample": 787200, "samples": 794400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u00-c07.mp3" }, "de-conversation-03-base17-segment-timed-u03-c00": { "id": "de-conversation-03-base17-segment-timed-u03-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85149, "plain": "Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 203520, "script_start_s": 0.24, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5228600188192065, "source_start_sample": 7680, "source_end_sample": 207360, "first_word_output_sample": 3840, "last_word_output_sample": 195840, "samples": 203040, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-timed-u03-c00.mp3" }, "en-conversation-05-base17-block-timed-u03-c00": { "id": "en-conversation-05-base17-block-timed-u03-c00", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 77649, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 23.12, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1102080, "script_start_s": 0.16, "script_end_s": 22.96, "all_asr_last_word_s": 22.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.1771345784162291, "source_start_sample": 3840, "source_end_sample": 1109280, "first_word_output_sample": 3840, "last_word_output_sample": 1098240, "samples": 1105440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u03-c00.mp3" }, "en-conversation-05-base17-block-timed-u03-c01": { "id": "en-conversation-05-base17-block-timed-u03-c01", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 77650, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 23.12, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1105920, "script_start_s": 0.0, "script_end_s": 23.04, "all_asr_last_word_s": 23.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u03-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.1536259968045033, "source_start_sample": 0, "source_end_sample": 1109760, "first_word_output_sample": 0, "last_word_output_sample": 1105920, "samples": 1113120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u03-c01.mp3" }, "en-conversation-05-base17-block-timed-u03-c02": { "id": "en-conversation-05-base17-block-timed-u03-c02", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 77651, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 23.12, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1102080, "script_start_s": 0.16, "script_end_s": 22.96, "all_asr_last_word_s": 22.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u03-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.23965640596296636, "source_start_sample": 3840, "source_end_sample": 1109280, "first_word_output_sample": 3840, "last_word_output_sample": 1098240, "samples": 1105440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u03-c02.mp3" }, "en-conversation-05-base17-block-timed-u03-c03": { "id": "en-conversation-05-base17-block-timed-u03-c03", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 77652, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 23.04, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1098240, "script_start_s": 0.16, "script_end_s": 22.88, "all_asr_last_word_s": 22.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u03-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.10390114865410699, "source_start_sample": 3840, "source_end_sample": 1105440, "first_word_output_sample": 3840, "last_word_output_sample": 1094400, "samples": 1101600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u03-c03.mp3" }, "en-conversation-05-base17-block-timed-u03-c04": { "id": "en-conversation-05-base17-block-timed-u03-c04", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 77653, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 23.12, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1105920, "script_start_s": 0.0, "script_end_s": 23.04, "all_asr_last_word_s": 23.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u03-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.16248271311929943, "source_start_sample": 0, "source_end_sample": 1109760, "first_word_output_sample": 0, "last_word_output_sample": 1105920, "samples": 1113120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u03-c04.mp3" }, "en-conversation-05-base17-block-timed-u03-c05": { "id": "en-conversation-05-base17-block-timed-u03-c05", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 77654, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 23.12, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1102080, "script_start_s": 0.0, "script_end_s": 22.96, "all_asr_last_word_s": 22.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u03-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.17753514582040733, "source_start_sample": 0, "source_end_sample": 1109280, "first_word_output_sample": 0, "last_word_output_sample": 1102080, "samples": 1109280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u03-c05.mp3" }, "en-conversation-05-base17-block-timed-u03-c06": { "id": "en-conversation-05-base17-block-timed-u03-c06", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 77655, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 23.04, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1098240, "script_start_s": 0.16, "script_end_s": 22.88, "all_asr_last_word_s": 22.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u03-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.13956112852214994, "source_start_sample": 3840, "source_end_sample": 1105440, "first_word_output_sample": 3840, "last_word_output_sample": 1094400, "samples": 1101600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u03-c06.mp3" }, "en-conversation-05-base17-block-timed-u03-c07": { "id": "en-conversation-05-base17-block-timed-u03-c07", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 77656, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 23.12, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1102080, "script_start_s": 0.0, "script_end_s": 22.96, "all_asr_last_word_s": 22.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u03-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.13554155698577486, "source_start_sample": 0, "source_end_sample": 1109280, "first_word_output_sample": 0, "last_word_output_sample": 1102080, "samples": 1109280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u03-c07.mp3" }, "en-conversation-02-base17-segment-timed-u06-c00": { "id": "en-conversation-02-base17-segment-timed-u06-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74197, "plain": "That is why the reference clips sit beside the output. You can compare for yourself.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 230400, "script_start_s": 0.24, "script_end_s": 4.8, "all_asr_last_word_s": 4.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.36619622873256386, "source_start_sample": 7680, "source_end_sample": 237600, "first_word_output_sample": 3840, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-timed-u06-c00.mp3" }, "de-conversation-01-base17-segment-timed-u07-c00": { "id": "de-conversation-01-base17-segment-timed-u07-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83213, "plain": "Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 13, "coverage": 0.9285714285714286, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 230400, "script_start_s": 0.24, "script_end_s": 4.8, "all_asr_last_word_s": 4.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07142857142857142, "source_identity_reward": 0.4300657127046441, "source_start_sample": 7680, "source_end_sample": 237600, "first_word_output_sample": 3840, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-timed-u07-c00.mp3" }, "de-conversation-02-base17-segment-timed-u01-c00": { "id": "de-conversation-02-base17-segment-timed-u01-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84117, "plain": "Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet.", "source_duration_s": 3.92, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 9, "coverage": 0.8181818181818182, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.18181818181818182, "source_identity_reward": 0.5787203621856052 }, "de-conversation-04-base17-segment-free-u00-c00": { "id": "de-conversation-04-base17-segment-free-u00-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86101, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 368640, "script_start_s": 0.24, "script_end_s": 7.68, "all_asr_last_word_s": 28.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 2.5, "source_identity_reward": 0.20983028340525664, "source_start_sample": 7680, "source_end_sample": 375840, "first_word_output_sample": 3840, "last_word_output_sample": 360960, "samples": 368160, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 22.17, "added_silence_s": 0.0, "duration_s": 7.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-free-u00-c00.mp3" }, "de-conversation-04-base17-segment-timed-u08-c00": { "id": "de-conversation-04-base17-segment-timed-u08-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.596198825588745, "source_start_sample": 0, "source_end_sample": 222720, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-timed-u08-c00.mp3" }, "en-conversation-04-base17-segment-timed-u05-c00": { "id": "en-conversation-04-base17-segment-timed-u05-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76181, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 218880, "script_start_s": 0.24, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4248773023927287, "source_start_sample": 7680, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 211200, "samples": 218400, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-timed-u05-c00.mp3" }, "en-conversation-02-base17-segment-timed-u03-c00": { "id": "en-conversation-02-base17-segment-timed-u03-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74149, "plain": "And does the second host interrupt with healthy skepticism, or just politely agree?", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 203520, "script_start_s": 0.0, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2571095459040763, "source_start_sample": 0, "source_end_sample": 207360, "first_word_output_sample": 0, "last_word_output_sample": 203520, "samples": 210720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-timed-u03-c00.mp3" }, "de-conversation-01-base17-segment-free-u05-c00": { "id": "de-conversation-01-base17-segment-free-u05-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83181, "plain": "Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 334080, "script_start_s": 0.0, "script_end_s": 6.96, "all_asr_last_word_s": 6.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6226884027586612, "source_start_sample": 0, "source_end_sample": 341280, "first_word_output_sample": 0, "last_word_output_sample": 334080, "samples": 341280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 22.89, "added_silence_s": 0.0, "duration_s": 7.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-free-u05-c00.mp3" }, "de-conversation-05-base17-block-free-u02-c00": { "id": "de-conversation-05-base17-block-free-u02-c00", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 87633, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 22.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07142857142857142, "source_identity_reward": 0.41145304819779355, "source_start_sample": 0, "source_end_sample": 963360, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 39.93, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u02-c00.mp3" }, "de-conversation-05-base17-block-free-u02-c01": { "id": "de-conversation-05-base17-block-free-u02-c01", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 87634, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 944640, "script_start_s": 0.16, "script_end_s": 19.68, "all_asr_last_word_s": 24.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.25, "source_identity_reward": 0.31874842351345334, "source_start_sample": 3840, "source_end_sample": 951840, "first_word_output_sample": 3840, "last_word_output_sample": 940800, "samples": 948000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.17, "added_silence_s": 0.0, "duration_s": 19.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u02-c01.mp3" }, "de-conversation-05-base17-block-free-u02-c02": { "id": "de-conversation-05-base17-block-free-u02-c02", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 87635, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 31.6, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 986880, "script_start_s": 0.24, "script_end_s": 20.56, "all_asr_last_word_s": 30.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.39285714285714285, "source_identity_reward": 0.17839465649123964, "source_start_sample": 7680, "source_end_sample": 994080, "first_word_output_sample": 3840, "last_word_output_sample": 979200, "samples": 986400, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 10.89, "added_silence_s": 0.0, "duration_s": 20.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u02-c02.mp3" }, "de-conversation-05-base17-block-free-u02-c03": { "id": "de-conversation-05-base17-block-free-u02-c03", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 87636, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 960000, "script_start_s": 0.24, "script_end_s": 20.0, "all_asr_last_word_s": 37.160000000000004 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.14285714285714285, "source_identity_reward": 0.29429851055995426, "source_start_sample": 7680, "source_end_sample": 967200, "first_word_output_sample": 3840, "last_word_output_sample": 952320, "samples": 959520, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 39.85, "added_silence_s": 0.0, "duration_s": 19.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u02-c03.mp3" }, "de-conversation-05-base17-block-free-u02-c04": { "id": "de-conversation-05-base17-block-free-u02-c04", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 87637, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 983040, "script_start_s": 0.24, "script_end_s": 20.48, "all_asr_last_word_s": 27.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.25, "source_identity_reward": 0.17068768264753437, "source_start_sample": 7680, "source_end_sample": 990240, "first_word_output_sample": 3840, "last_word_output_sample": 975360, "samples": 982560, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 39.37, "added_silence_s": 0.0, "duration_s": 20.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u02-c04.mp3" }, "de-conversation-05-base17-block-free-u02-c05": { "id": "de-conversation-05-base17-block-free-u02-c05", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 87638, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 952320, "script_start_s": 0.0, "script_end_s": 19.84, "all_asr_last_word_s": 44.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.32142857142857145, "source_identity_reward": 0.14716809367682662, "source_start_sample": 0, "source_end_sample": 959520, "first_word_output_sample": 0, "last_word_output_sample": 952320, "samples": 959520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 40.01, "added_silence_s": 0.0, "duration_s": 19.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u02-c05.mp3" }, "de-conversation-05-base17-block-free-u02-c06": { "id": "de-conversation-05-base17-block-free-u02-c06", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 87639, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1025280, "script_start_s": 0.08, "script_end_s": 21.36, "all_asr_last_word_s": 28.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.19642857142857142, "source_identity_reward": 0.2007949666455138, "source_start_sample": 0, "source_end_sample": 1032480, "first_word_output_sample": 3840, "last_word_output_sample": 1025280, "samples": 1032480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.49, "added_silence_s": 0.0, "duration_s": 21.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u02-c06.mp3" }, "de-conversation-05-base17-block-free-u02-c07": { "id": "de-conversation-05-base17-block-free-u02-c07", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 87640, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1098240, "script_start_s": 0.16, "script_end_s": 22.88, "all_asr_last_word_s": 26.560000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.10714285714285714, "source_identity_reward": 0.23574905798715565, "source_start_sample": 3840, "source_end_sample": 1105440, "first_word_output_sample": 3840, "last_word_output_sample": 1094400, "samples": 1101600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 36.97, "added_silence_s": 0.0, "duration_s": 22.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u02-c07.mp3" }, "de-conversation-04-base17-segment-free-u01-c00": { "id": "de-conversation-04-base17-segment-free-u01-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86117, "plain": "Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde?", "source_duration_s": 16.64, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 299520, "script_start_s": 0.16, "script_end_s": 6.24, "all_asr_last_word_s": 14.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0909090909090908, "source_identity_reward": 0.2874170928272825, "source_start_sample": 3840, "source_end_sample": 306720, "first_word_output_sample": 3840, "last_word_output_sample": 295680, "samples": 302880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 10.25, "added_silence_s": 0.0, "duration_s": 6.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-free-u01-c00.mp3" }, "de-conversation-03-base17-segment-free-u02-c00": { "id": "de-conversation-03-base17-segment-free-u02-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85133, "plain": "Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 337920, "script_start_s": 0.0, "script_end_s": 7.04, "all_asr_last_word_s": 25.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0, "source_identity_reward": 0.3367345464268815, "source_start_sample": 0, "source_end_sample": 345120, "first_word_output_sample": 0, "last_word_output_sample": 337920, "samples": 345120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 22.81, "added_silence_s": 0.0, "duration_s": 7.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-free-u02-c00.mp3" }, "en-conversation-05-base17-segment-free-u10-c00": { "id": "en-conversation-05-base17-segment-free-u10-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 22.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 480000, "script_start_s": 0.0, "script_end_s": 10.0, "all_asr_last_word_s": 14.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.35714285714285715, "source_identity_reward": 0.28759405591560083, "source_start_sample": 0, "source_end_sample": 487200, "first_word_output_sample": 0, "last_word_output_sample": 480000, "samples": 487200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 12.49, "added_silence_s": 0.0, "duration_s": 10.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-free-u10-c00.mp3" }, "en-conversation-04-base17-segment-free-u05-c00": { "id": "en-conversation-04-base17-segment-free-u05-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76181, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 291840, "script_start_s": 0.16, "script_end_s": 6.08, "all_asr_last_word_s": 6.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.15112957839035104, "source_start_sample": 3840, "source_end_sample": 299040, "first_word_output_sample": 3840, "last_word_output_sample": 288000, "samples": 295200, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.77, "added_silence_s": 0.0, "duration_s": 6.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-free-u05-c00.mp3" }, "en-conversation-02-base17-block-free-u01-c00": { "id": "en-conversation-02-base17-block-free-u01-c00", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 74617, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 70, "matched_tokens": 9, "coverage": 0.12857142857142856, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.8714285714285714, "source_identity_reward": 0.01719566432937761 }, "en-conversation-02-base17-block-free-u01-c01": { "id": "en-conversation-02-base17-block-free-u01-c01", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 74618, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 70, "matched_tokens": 65, "coverage": 0.9285714285714286, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07142857142857142, "source_identity_reward": 0.10116968090916834 }, "en-conversation-02-base17-block-free-u01-c02": { "id": "en-conversation-02-base17-block-free-u01-c02", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 74619, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1004160, "script_start_s": 0.24, "script_end_s": 20.92, "all_asr_last_word_s": 28.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2714285714285714, "source_identity_reward": 0.11665496315765292, "source_start_sample": 7680, "source_end_sample": 1011360, "first_word_output_sample": 3840, "last_word_output_sample": 996480, "samples": 1003680, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 38.93, "added_silence_s": 0.0, "duration_s": 20.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u01-c02.mp3" }, "en-conversation-02-base17-block-free-u01-c03": { "id": "en-conversation-02-base17-block-free-u01-c03", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 74620, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 70, "matched_tokens": 38, "coverage": 0.5428571428571428, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.45714285714285713, "source_identity_reward": 0.08774958616136407 }, "en-conversation-02-base17-block-free-u01-c04": { "id": "en-conversation-02-base17-block-free-u01-c04", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 74621, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 960000, "script_start_s": 0.0, "script_end_s": 20.0, "all_asr_last_word_s": 46.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4, "source_identity_reward": 0.1290654605149153, "source_start_sample": 0, "source_end_sample": 967200, "first_word_output_sample": 0, "last_word_output_sample": 960000, "samples": 967200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 39.85, "added_silence_s": 0.0, "duration_s": 20.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u01-c04.mp3" }, "en-conversation-02-base17-block-free-u01-c05": { "id": "en-conversation-02-base17-block-free-u01-c05", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 74622, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1052160, "script_start_s": 0.16, "script_end_s": 21.92, "all_asr_last_word_s": 39.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2714285714285714, "source_identity_reward": 0.1430429496929049, "source_start_sample": 3840, "source_end_sample": 1059360, "first_word_output_sample": 3840, "last_word_output_sample": 1048320, "samples": 1055520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 37.93, "added_silence_s": 0.0, "duration_s": 21.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u01-c05.mp3" }, "en-conversation-02-base17-block-free-u01-c06": { "id": "en-conversation-02-base17-block-free-u01-c06", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 74623, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1029120, "script_start_s": 0.16, "script_end_s": 21.44, "all_asr_last_word_s": 48.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4857142857142857, "source_identity_reward": 0.14730197405600956, "source_start_sample": 3840, "source_end_sample": 1036320, "first_word_output_sample": 3840, "last_word_output_sample": 1025280, "samples": 1032480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 38.41, "added_silence_s": 0.0, "duration_s": 21.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u01-c06.mp3" }, "en-conversation-02-base17-block-free-u01-c07": { "id": "en-conversation-02-base17-block-free-u01-c07", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 74624, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 983040, "script_start_s": 0.24, "script_end_s": 20.48, "all_asr_last_word_s": 41.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3142857142857143, "source_identity_reward": 0.12893052642234762, "source_start_sample": 7680, "source_end_sample": 990240, "first_word_output_sample": 3840, "last_word_output_sample": 975360, "samples": 982560, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 39.37, "added_silence_s": 0.0, "duration_s": 20.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u01-c07.mp3" }, "de-conversation-04-base17-segment-free-u02-c00": { "id": "de-conversation-04-base17-segment-free-u02-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86133, "plain": "Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden.", "source_duration_s": 14.96, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 10, "coverage": 0.6666666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 314880, "script_start_s": 0.16, "script_end_s": 6.56, "all_asr_last_word_s": 6.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3333333333333333, "source_identity_reward": 0.6053814708469705, "source_start_sample": 3840, "source_end_sample": 322080, "first_word_output_sample": 3840, "last_word_output_sample": 311040, "samples": 318240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 8.25, "added_silence_s": 0.0, "duration_s": 6.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-free-u02-c00.mp3" }, "en-conversation-04-base17-segment-free-u06-c00": { "id": "en-conversation-04-base17-segment-free-u06-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76197, "plain": "That is the idea: control the intention, measure the result, and listen carefully.", "source_duration_s": 9.2, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 234240, "script_start_s": 0.0, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.49287238608133704, "source_start_sample": 0, "source_end_sample": 241440, "first_word_output_sample": 0, "last_word_output_sample": 234240, "samples": 241440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.17, "added_silence_s": 0.0, "duration_s": 5.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-free-u06-c00.mp3" }, "en-conversation-04-base17-segment-timed-u01-c00": { "id": "en-conversation-04-base17-segment-timed-u01-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76117, "plain": "You are kidding. Do you mean it lands on an exact millisecond?", "source_duration_s": 4.0, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 184320, "script_start_s": 0.24, "script_end_s": 3.84, "all_asr_last_word_s": 3.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1467097033840355, "source_start_sample": 7680, "source_end_sample": 191520, "first_word_output_sample": 3840, "last_word_output_sample": 176640, "samples": 183840, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 3.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-timed-u01-c00.mp3" }, "en-conversation-04-base17-block-timed-u02-c00": { "id": "en-conversation-04-base17-block-timed-u02-c00", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 76633, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 460800, "script_start_s": 0.16, "script_end_s": 9.6, "all_asr_last_word_s": 9.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.20976251061126022, "source_start_sample": 3840, "source_end_sample": 464640, "first_word_output_sample": 3840, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u02-c00.mp3" }, "en-conversation-04-base17-block-timed-u02-c01": { "id": "en-conversation-04-base17-block-timed-u02-c01", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 76634, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 460800, "script_start_s": 0.16, "script_end_s": 9.6, "all_asr_last_word_s": 9.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.12287803251886922, "source_start_sample": 3840, "source_end_sample": 464640, "first_word_output_sample": 3840, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u02-c01.mp3" }, "en-conversation-04-base17-block-timed-u02-c02": { "id": "en-conversation-04-base17-block-timed-u02-c02", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 76635, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 456960, "script_start_s": 0.16, "script_end_s": 9.52, "all_asr_last_word_s": 9.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.22254823827298645, "source_start_sample": 3840, "source_end_sample": 464160, "first_word_output_sample": 3840, "last_word_output_sample": 453120, "samples": 460320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u02-c02.mp3" }, "en-conversation-04-base17-block-timed-u02-c03": { "id": "en-conversation-04-base17-block-timed-u02-c03", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 76636, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.76, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 28, "coverage": 0.9655172413793104, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 464640, "script_start_s": 0.16, "script_end_s": 9.68, "all_asr_last_word_s": 9.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.034482758620689655, "source_identity_reward": 0.20611428964797698, "source_start_sample": 3840, "source_end_sample": 468480, "first_word_output_sample": 3840, "last_word_output_sample": 460800, "samples": 468000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u02-c03.mp3" }, "en-conversation-04-base17-block-timed-u02-c04": { "id": "en-conversation-04-base17-block-timed-u02-c04", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 76637, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 460800, "script_start_s": 0.0, "script_end_s": 9.6, "all_asr_last_word_s": 9.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.13684875583023198, "source_start_sample": 0, "source_end_sample": 464640, "first_word_output_sample": 0, "last_word_output_sample": 460800, "samples": 468000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u02-c04.mp3" }, "en-conversation-04-base17-block-timed-u02-c05": { "id": "en-conversation-04-base17-block-timed-u02-c05", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 76638, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 456960, "script_start_s": 0.0, "script_end_s": 9.52, "all_asr_last_word_s": 9.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.09227886302315536, "source_start_sample": 0, "source_end_sample": 464160, "first_word_output_sample": 0, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u02-c05.mp3" }, "en-conversation-04-base17-block-timed-u02-c06": { "id": "en-conversation-04-base17-block-timed-u02-c06", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 76639, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 460800, "script_start_s": 0.0, "script_end_s": 9.6, "all_asr_last_word_s": 9.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1477784389460689, "source_start_sample": 0, "source_end_sample": 464640, "first_word_output_sample": 0, "last_word_output_sample": 460800, "samples": 468000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u02-c06.mp3" }, "en-conversation-04-base17-block-timed-u02-c07": { "id": "en-conversation-04-base17-block-timed-u02-c07", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 76640, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 460800, "script_start_s": 0.16, "script_end_s": 9.6, "all_asr_last_word_s": 9.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.19613245463533543, "source_start_sample": 3840, "source_end_sample": 464640, "first_word_output_sample": 3840, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u02-c07.mp3" }, "de-conversation-02-base17-segment-timed-u10-c00": { "id": "de-conversation-02-base17-segment-timed-u10-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 5.68, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 268800, "script_start_s": 0.0, "script_end_s": 5.6, "all_asr_last_word_s": 5.6000000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6222886334241176, "source_start_sample": 0, "source_end_sample": 272640, "first_word_output_sample": 0, "last_word_output_sample": 268800, "samples": 276000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-timed-u10-c00.mp3" }, "de-conversation-03-base17-block-timed-u00-c00": { "id": "de-conversation-03-base17-block-timed-u00-c00", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 85601, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.6, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 840960, "script_start_s": 0.0, "script_end_s": 17.52, "all_asr_last_word_s": 17.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2246767548620714, "source_start_sample": 0, "source_end_sample": 844800, "first_word_output_sample": 0, "last_word_output_sample": 840960, "samples": 848160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-timed-u00-c00.mp3" }, "de-conversation-03-base17-block-timed-u00-c01": { "id": "de-conversation-03-base17-block-timed-u00-c01", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 85602, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.6, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 837120, "script_start_s": 0.16, "script_end_s": 17.44, "all_asr_last_word_s": 17.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.238535211609319, "source_start_sample": 3840, "source_end_sample": 844320, "first_word_output_sample": 3840, "last_word_output_sample": 833280, "samples": 840480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-timed-u00-c01.mp3" }, "de-conversation-03-base17-block-timed-u00-c02": { "id": "de-conversation-03-base17-block-timed-u00-c02", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 85603, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.52, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 837120, "script_start_s": 0.0, "script_end_s": 17.44, "all_asr_last_word_s": 17.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3006020970822558, "source_start_sample": 0, "source_end_sample": 840960, "first_word_output_sample": 0, "last_word_output_sample": 837120, "samples": 844320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-timed-u00-c02.mp3" }, "de-conversation-03-base17-block-timed-u00-c03": { "id": "de-conversation-03-base17-block-timed-u00-c03", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 85604, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.6, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 840960, "script_start_s": 0.16, "script_end_s": 17.52, "all_asr_last_word_s": 17.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2455311559760668, "source_start_sample": 3840, "source_end_sample": 844800, "first_word_output_sample": 3840, "last_word_output_sample": 837120, "samples": 844320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-timed-u00-c03.mp3" }, "de-conversation-03-base17-block-timed-u00-c04": { "id": "de-conversation-03-base17-block-timed-u00-c04", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 85605, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.52, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 837120, "script_start_s": 0.0, "script_end_s": 17.44, "all_asr_last_word_s": 17.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.19218047728628943, "source_start_sample": 0, "source_end_sample": 840960, "first_word_output_sample": 0, "last_word_output_sample": 837120, "samples": 844320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-timed-u00-c04.mp3" }, "de-conversation-03-base17-block-timed-u00-c05": { "id": "de-conversation-03-base17-block-timed-u00-c05", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 85606, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.6, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 840960, "script_start_s": 0.0, "script_end_s": 17.52, "all_asr_last_word_s": 17.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2677253759829114, "source_start_sample": 0, "source_end_sample": 844800, "first_word_output_sample": 0, "last_word_output_sample": 840960, "samples": 848160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-timed-u00-c05.mp3" }, "de-conversation-03-base17-block-timed-u00-c06": { "id": "de-conversation-03-base17-block-timed-u00-c06", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 85607, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.52, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 837120, "script_start_s": 0.0, "script_end_s": 17.44, "all_asr_last_word_s": 17.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.40022778198010933, "source_start_sample": 0, "source_end_sample": 840960, "first_word_output_sample": 0, "last_word_output_sample": 837120, "samples": 844320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-timed-u00-c06.mp3" }, "de-conversation-03-base17-block-timed-u00-c07": { "id": "de-conversation-03-base17-block-timed-u00-c07", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 85608, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.44, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 48, "coverage": 0.9795918367346939, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02040816326530612, "source_identity_reward": 0.22137557111629402 }, "de-conversation-02-base17-segment-free-u11-c00": { "id": "de-conversation-02-base17-segment-free-u11-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 17.84, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 15360, "last_word_sample": 357120, "script_start_s": 0.32, "script_end_s": 7.44, "all_asr_last_word_s": 7.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6210427123422229, "source_start_sample": 11520, "source_end_sample": 364320, "first_word_output_sample": 3840, "last_word_output_sample": 345600, "samples": 352800, "sample_rate": 48000, "head_removed_s": 0.24, "tail_removed_s": 10.25, "added_silence_s": 0.0, "duration_s": 7.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-free-u11-c00.mp3" }, "en-conversation-02-base17-segment-free-u08-c00": { "id": "en-conversation-02-base17-segment-free-u08-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 334080, "script_start_s": 0.16, "script_end_s": 6.96, "all_asr_last_word_s": 29.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.2142857142857142, "source_identity_reward": 0.10475304890585606, "source_start_sample": 3840, "source_end_sample": 341280, "first_word_output_sample": 3840, "last_word_output_sample": 330240, "samples": 337440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 22.89, "added_silence_s": 0.0, "duration_s": 7.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-free-u08-c00.mp3" }, "de-conversation-05-base17-segment-timed-u10-c00": { "id": "de-conversation-05-base17-segment-timed-u10-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 5.68, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 264960, "script_start_s": 0.0, "script_end_s": 5.52, "all_asr_last_word_s": 5.5200000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6234338558254788, "source_start_sample": 0, "source_end_sample": 272160, "first_word_output_sample": 0, "last_word_output_sample": 264960, "samples": 272160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-timed-u10-c00.mp3" }, "de-conversation-01-base17-segment-timed-u03-c00": { "id": "de-conversation-01-base17-segment-timed-u03-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83149, "plain": "Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 3.92, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 184320, "script_start_s": 0.16, "script_end_s": 3.84, "all_asr_last_word_s": 3.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5526991032984413, "source_start_sample": 3840, "source_end_sample": 188160, "first_word_output_sample": 3840, "last_word_output_sample": 180480, "samples": 187680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-timed-u03-c00.mp3" }, "en-conversation-01-base17-block-free-u02-c00": { "id": "en-conversation-01-base17-block-free-u02-c00", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 73633, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 844800, "script_start_s": 0.16, "script_end_s": 17.6, "all_asr_last_word_s": 19.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.10909090909090909, "source_identity_reward": 0.2045506127902492, "source_start_sample": 3840, "source_end_sample": 852000, "first_word_output_sample": 3840, "last_word_output_sample": 840960, "samples": 848160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.53, "added_silence_s": 0.0, "duration_s": 17.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u02-c00.mp3" }, "en-conversation-01-base17-block-free-u02-c01": { "id": "en-conversation-01-base17-block-free-u02-c01", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 73634, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 19.36, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 898560, "script_start_s": 0.24, "script_end_s": 18.72, "all_asr_last_word_s": 18.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.11958043804971234, "source_start_sample": 7680, "source_end_sample": 905760, "first_word_output_sample": 3840, "last_word_output_sample": 890880, "samples": 898080, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.49, "added_silence_s": 0.0, "duration_s": 18.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u02-c01.mp3" }, "en-conversation-01-base17-block-free-u02-c02": { "id": "en-conversation-01-base17-block-free-u02-c02", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 73635, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 19.76, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 921600, "script_start_s": 0.16, "script_end_s": 19.2, "all_asr_last_word_s": 19.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.11074159027698552, "source_start_sample": 3840, "source_end_sample": 928800, "first_word_output_sample": 3840, "last_word_output_sample": 917760, "samples": 924960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.41, "added_silence_s": 0.0, "duration_s": 19.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u02-c02.mp3" }, "en-conversation-01-base17-block-free-u02-c03": { "id": "en-conversation-01-base17-block-free-u02-c03", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 73636, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 913920, "script_start_s": 0.24, "script_end_s": 19.04, "all_asr_last_word_s": 44.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4, "source_identity_reward": 0.06525621675683296, "source_start_sample": 7680, "source_end_sample": 921120, "first_word_output_sample": 3840, "last_word_output_sample": 906240, "samples": 913440, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 40.81, "added_silence_s": 0.0, "duration_s": 19.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u02-c03.mp3" }, "en-conversation-01-base17-block-free-u02-c04": { "id": "en-conversation-01-base17-block-free-u02-c04", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 73637, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 27.68, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 852480, "script_start_s": 0.16, "script_end_s": 17.76, "all_asr_last_word_s": 17.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.13470464856205616, "source_start_sample": 3840, "source_end_sample": 859680, "first_word_output_sample": 3840, "last_word_output_sample": 848640, "samples": 855840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 9.77, "added_silence_s": 0.0, "duration_s": 17.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u02-c04.mp3" }, "en-conversation-01-base17-block-free-u02-c05": { "id": "en-conversation-01-base17-block-free-u02-c05", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 73638, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 19.52, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 917760, "script_start_s": 0.16, "script_end_s": 19.12, "all_asr_last_word_s": 19.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.10158057573784589, "source_start_sample": 3840, "source_end_sample": 924960, "first_word_output_sample": 3840, "last_word_output_sample": 913920, "samples": 921120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.25, "added_silence_s": 0.0, "duration_s": 19.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u02-c05.mp3" }, "en-conversation-01-base17-block-free-u02-c06": { "id": "en-conversation-01-base17-block-free-u02-c06", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 73639, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 814080, "script_start_s": 0.16, "script_end_s": 16.96, "all_asr_last_word_s": 52.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.38181818181818183, "source_identity_reward": 0.11476812701394047, "source_start_sample": 3840, "source_end_sample": 821280, "first_word_output_sample": 3840, "last_word_output_sample": 810240, "samples": 817440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 42.89, "added_silence_s": 0.0, "duration_s": 17.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u02-c06.mp3" }, "en-conversation-01-base17-block-free-u02-c07": { "id": "en-conversation-01-base17-block-free-u02-c07", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 73640, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1655040, "script_start_s": 0.0, "script_end_s": 34.48, "all_asr_last_word_s": 34.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.14545454545454545, "source_identity_reward": 0.19359297248272903, "source_start_sample": 0, "source_end_sample": 1662240, "first_word_output_sample": 0, "last_word_output_sample": 1655040, "samples": 1662240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 25.37, "added_silence_s": 0.0, "duration_s": 34.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u02-c07.mp3" }, "de-conversation-03-base17-segment-free-u11-c00": { "id": "de-conversation-03-base17-segment-free-u11-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 341760, "script_start_s": 0.24, "script_end_s": 7.12, "all_asr_last_word_s": 7.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6342478186043284, "source_start_sample": 7680, "source_end_sample": 348960, "first_word_output_sample": 3840, "last_word_output_sample": 334080, "samples": 341280, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 22.73, "added_silence_s": 0.0, "duration_s": 7.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-free-u11-c00.mp3" }, "de-conversation-05-base17-segment-free-u12-c00": { "id": "de-conversation-05-base17-segment-free-u12-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87293, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen.", "source_duration_s": 9.6, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 349440, "script_start_s": 0.08, "script_end_s": 7.28, "all_asr_last_word_s": 7.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u12-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7041685562289899, "source_start_sample": 0, "source_end_sample": 356640, "first_word_output_sample": 3840, "last_word_output_sample": 349440, "samples": 356640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.17, "added_silence_s": 0.0, "duration_s": 7.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u12-c00.mp3" }, "de-conversation-03-base17-segment-free-u07-c00": { "id": "de-conversation-03-base17-segment-free-u07-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85213, "plain": "Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 314880, "script_start_s": 0.24, "script_end_s": 6.56, "all_asr_last_word_s": 29.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5454545454545454, "source_identity_reward": 0.2709248304650617, "source_start_sample": 7680, "source_end_sample": 322080, "first_word_output_sample": 3840, "last_word_output_sample": 307200, "samples": 314400, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 23.29, "added_silence_s": 0.0, "duration_s": 6.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-free-u07-c00.mp3" }, "en-conversation-02-base17-segment-free-u09-c00": { "id": "en-conversation-02-base17-segment-free-u09-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 9.52, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 453120, "script_start_s": 0.24, "script_end_s": 9.44, "all_asr_last_word_s": 9.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.24978602483621887, "source_start_sample": 7680, "source_end_sample": 456960, "first_word_output_sample": 3840, "last_word_output_sample": 445440, "samples": 452640, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-free-u09-c00.mp3" }, "en-conversation-02-base17-segment-free-u04-c00": { "id": "en-conversation-02-base17-segment-free-u04-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74165, "plain": "Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 9.04, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 15360, "last_word_sample": 430080, "script_start_s": 0.32, "script_end_s": 8.96, "all_asr_last_word_s": 8.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.37283060527355244, "source_start_sample": 11520, "source_end_sample": 433920, "first_word_output_sample": 3840, "last_word_output_sample": 418560, "samples": 425760, "sample_rate": 48000, "head_removed_s": 0.24, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 8.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-free-u04-c00.mp3" }, "en-conversation-05-base17-block-free-u01-c00": { "id": "en-conversation-05-base17-block-free-u01-c00", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 77617, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 59, "matched_tokens": 58, "coverage": 0.9830508474576272, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 888960, "script_start_s": 0.16, "script_end_s": 18.52, "all_asr_last_word_s": 34.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1016949152542373, "source_identity_reward": 0.2136937650439945, "source_start_sample": 3840, "source_end_sample": 896160, "first_word_output_sample": 3840, "last_word_output_sample": 885120, "samples": 892320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 41.33, "added_silence_s": 0.0, "duration_s": 18.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u01-c00.mp3" }, "en-conversation-05-base17-block-free-u01-c01": { "id": "en-conversation-05-base17-block-free-u01-c01", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 77618, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 839040, "script_start_s": 0.0, "script_end_s": 17.48, "all_asr_last_word_s": 37.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0847457627118644, "source_identity_reward": 0.16506028937226636, "source_start_sample": 0, "source_end_sample": 846240, "first_word_output_sample": 0, "last_word_output_sample": 839040, "samples": 846240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 42.37, "added_silence_s": 0.0, "duration_s": 17.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u01-c01.mp3" }, "en-conversation-05-base17-block-free-u01-c02": { "id": "en-conversation-05-base17-block-free-u01-c02", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 77619, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 59, "matched_tokens": 53, "coverage": 0.8983050847457628, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 929280, "script_start_s": 0.0, "script_end_s": 19.36, "all_asr_last_word_s": 39.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3559322033898305, "source_identity_reward": 0.11539937827912633, "source_start_sample": 0, "source_end_sample": 936480, "first_word_output_sample": 0, "last_word_output_sample": 929280, "samples": 936480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 40.49, "added_silence_s": 0.0, "duration_s": 19.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u01-c02.mp3" }, "en-conversation-05-base17-block-free-u01-c03": { "id": "en-conversation-05-base17-block-free-u01-c03", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 77620, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 59, "matched_tokens": 56, "coverage": 0.9491525423728814, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 867840, "script_start_s": 0.16, "script_end_s": 18.08, "all_asr_last_word_s": 18.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05084745762711865, "source_identity_reward": 0.15544035507661524, "source_start_sample": 3840, "source_end_sample": 875040, "first_word_output_sample": 3840, "last_word_output_sample": 864000, "samples": 871200, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 41.77, "added_silence_s": 0.0, "duration_s": 18.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u01-c03.mp3" }, "en-conversation-05-base17-block-free-u01-c04": { "id": "en-conversation-05-base17-block-free-u01-c04", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 77621, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 59, "matched_tokens": 57, "coverage": 0.9661016949152542, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 821760, "script_start_s": 0.16, "script_end_s": 17.12, "all_asr_last_word_s": 46.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2542372881355932, "source_identity_reward": 0.1963963390863333, "source_start_sample": 3840, "source_end_sample": 828960, "first_word_output_sample": 3840, "last_word_output_sample": 817920, "samples": 825120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 42.73, "added_silence_s": 0.0, "duration_s": 17.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u01-c04.mp3" }, "en-conversation-05-base17-block-free-u01-c05": { "id": "en-conversation-05-base17-block-free-u01-c05", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 77622, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 873600, "script_start_s": 0.16, "script_end_s": 18.2, "all_asr_last_word_s": 53.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2711864406779661, "source_identity_reward": 0.1554288578554973, "source_start_sample": 3840, "source_end_sample": 880800, "first_word_output_sample": 3840, "last_word_output_sample": 869760, "samples": 876960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 41.65, "added_silence_s": 0.0, "duration_s": 18.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u01-c05.mp3" }, "en-conversation-05-base17-block-free-u01-c06": { "id": "en-conversation-05-base17-block-free-u01-c06", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 77623, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 59, "matched_tokens": 54, "coverage": 0.9152542372881356, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 817920, "script_start_s": 0.0, "script_end_s": 17.04, "all_asr_last_word_s": 59.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3220338983050847, "source_identity_reward": 0.14683745086525998, "source_start_sample": 0, "source_end_sample": 825120, "first_word_output_sample": 0, "last_word_output_sample": 817920, "samples": 825120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 42.81, "added_silence_s": 0.0, "duration_s": 17.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u01-c06.mp3" }, "en-conversation-05-base17-block-free-u01-c07": { "id": "en-conversation-05-base17-block-free-u01-c07", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 77624, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 18.96, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 57, "coverage": 0.9661016949152542, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 871680, "script_start_s": 0.0, "script_end_s": 18.16, "all_asr_last_word_s": 18.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03389830508474576, "source_identity_reward": 0.21495032882024517, "source_start_sample": 0, "source_end_sample": 878880, "first_word_output_sample": 0, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.65, "added_silence_s": 0.0, "duration_s": 18.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u01-c07.mp3" }, "de-conversation-03-base17-segment-timed-u09-c00": { "id": "de-conversation-03-base17-segment-timed-u09-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 234240, "script_start_s": 0.0, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6760227915919346, "source_start_sample": 0, "source_end_sample": 238080, "first_word_output_sample": 0, "last_word_output_sample": 234240, "samples": 241440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-timed-u09-c00.mp3" }, "en-conversation-01-base17-segment-timed-u09-c00": { "id": "en-conversation-01-base17-segment-timed-u09-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 4.0, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 188160, "script_start_s": 0.16, "script_end_s": 3.92, "all_asr_last_word_s": 3.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.09641617177130933, "source_start_sample": 3840, "source_end_sample": 192000, "first_word_output_sample": 3840, "last_word_output_sample": 184320, "samples": 191520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-timed-u09-c00.mp3" }, "de-conversation-02-base17-segment-timed-u03-c00": { "id": "de-conversation-02-base17-segment-timed-u03-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84149, "plain": "Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen?", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 203520, "script_start_s": 0.0, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7029570883330856, "source_start_sample": 0, "source_end_sample": 207360, "first_word_output_sample": 0, "last_word_output_sample": 203520, "samples": 210720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-timed-u03-c00.mp3" }, "de-conversation-01-base17-segment-timed-u06-c00": { "id": "de-conversation-01-base17-segment-timed-u06-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83197, "plain": "Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 222720, "script_start_s": 0.0, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5403811970063449, "source_start_sample": 0, "source_end_sample": 226560, "first_word_output_sample": 0, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-timed-u06-c00.mp3" }, "en-conversation-03-base17-segment-free-u08-c00": { "id": "en-conversation-03-base17-segment-free-u08-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 288000, "script_start_s": 0.16, "script_end_s": 6.0, "all_asr_last_word_s": 6.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1579033652146436, "source_start_sample": 3840, "source_end_sample": 295200, "first_word_output_sample": 3840, "last_word_output_sample": 284160, "samples": 291360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.85, "added_silence_s": 0.0, "duration_s": 6.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-free-u08-c00.mp3" }, "de-conversation-01-base17-block-timed-u02-c00": { "id": "de-conversation-01-base17-block-timed-u02-c00", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 83633, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 956160, "script_start_s": 0.16, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3117939649686027, "source_start_sample": 3840, "source_end_sample": 960000, "first_word_output_sample": 3840, "last_word_output_sample": 952320, "samples": 959520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 19.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u02-c00.mp3" }, "de-conversation-01-base17-block-timed-u02-c01": { "id": "de-conversation-01-base17-block-timed-u02-c01", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 83634, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 952320, "script_start_s": 0.16, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.32754543239824296, "source_start_sample": 3840, "source_end_sample": 959520, "first_word_output_sample": 3840, "last_word_output_sample": 948480, "samples": 955680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u02-c01.mp3" }, "de-conversation-01-base17-block-timed-u02-c02": { "id": "de-conversation-01-base17-block-timed-u02-c02", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 83635, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 54, "coverage": 0.9642857142857143, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 952320, "script_start_s": 0.24, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.3074189045132111, "source_start_sample": 7680, "source_end_sample": 959520, "first_word_output_sample": 3840, "last_word_output_sample": 944640, "samples": 951840, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u02-c02.mp3" }, "de-conversation-01-base17-block-timed-u02-c03": { "id": "de-conversation-01-base17-block-timed-u02-c03", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 83636, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 952320, "script_start_s": 0.0, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2058928302264589, "source_start_sample": 0, "source_end_sample": 959520, "first_word_output_sample": 0, "last_word_output_sample": 952320, "samples": 959520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u02-c03.mp3" }, "de-conversation-01-base17-block-timed-u02-c04": { "id": "de-conversation-01-base17-block-timed-u02-c04", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 83637, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.08, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 960000, "script_start_s": 0.16, "script_end_s": 20.0, "all_asr_last_word_s": 20.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.27461035530349226, "source_start_sample": 3840, "source_end_sample": 963840, "first_word_output_sample": 3840, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u02-c04.mp3" }, "de-conversation-01-base17-block-timed-u02-c05": { "id": "de-conversation-01-base17-block-timed-u02-c05", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 83638, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 952320, "script_start_s": 0.16, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.22871728896175145, "source_start_sample": 3840, "source_end_sample": 959520, "first_word_output_sample": 3840, "last_word_output_sample": 948480, "samples": 955680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u02-c05.mp3" }, "de-conversation-01-base17-block-timed-u02-c06": { "id": "de-conversation-01-base17-block-timed-u02-c06", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 83639, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 952320, "script_start_s": 0.16, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.017857142857142856, "source_identity_reward": 0.2726162441881466, "source_start_sample": 3840, "source_end_sample": 959520, "first_word_output_sample": 3840, "last_word_output_sample": 948480, "samples": 955680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u02-c06.mp3" }, "de-conversation-01-base17-block-timed-u02-c07": { "id": "de-conversation-01-base17-block-timed-u02-c07", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 83640, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 54, "coverage": 0.9642857142857143, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 952320, "script_start_s": 0.0, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.16655347118541663, "source_start_sample": 0, "source_end_sample": 959520, "first_word_output_sample": 0, "last_word_output_sample": 952320, "samples": 959520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u02-c07.mp3" }, "en-conversation-03-base17-segment-free-u02-c00": { "id": "en-conversation-03-base17-segment-free-u02-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75133, "plain": "Yes. The model sees speaker-labelled references and, when available, earlier dialogue.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 11, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 291840, "script_start_s": 0.0, "script_end_s": 6.08, "all_asr_last_word_s": 16.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.5833333333333333, "source_identity_reward": 0.18975108773183447, "source_start_sample": 0, "source_end_sample": 299040, "first_word_output_sample": 0, "last_word_output_sample": 291840, "samples": 299040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.77, "added_silence_s": 0.0, "duration_s": 6.23, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-free-u02-c00.mp3" }, "de-conversation-05-base17-block-free-u00-c00": { "id": "de-conversation-05-base17-block-free-u00-c00", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 87601, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 46, "matched_tokens": 45, "coverage": 0.9782608695652174, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 948480, "script_start_s": 0.08, "script_end_s": 19.76, "all_asr_last_word_s": 58.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.13043478260869565, "source_identity_reward": 0.3109478443659162, "source_start_sample": 0, "source_end_sample": 955680, "first_word_output_sample": 3840, "last_word_output_sample": 948480, "samples": 955680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 40.09, "added_silence_s": 0.0, "duration_s": 19.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u00-c00.mp3" }, "de-conversation-05-base17-block-free-u00-c01": { "id": "de-conversation-05-base17-block-free-u00-c01", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 87602, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 23.76, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 45, "coverage": 0.9782608695652174, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1044480, "script_start_s": 0.0, "script_end_s": 21.76, "all_asr_last_word_s": 21.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.021739130434782608, "source_identity_reward": 0.26637058641259387, "source_start_sample": 0, "source_end_sample": 1051680, "first_word_output_sample": 0, "last_word_output_sample": 1044480, "samples": 1051680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.85, "added_silence_s": 0.0, "duration_s": 21.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u00-c01.mp3" }, "de-conversation-05-base17-block-free-u00-c02": { "id": "de-conversation-05-base17-block-free-u00-c02", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 87603, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 30.8, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 45, "coverage": 0.9782608695652174, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 990720, "script_start_s": 0.16, "script_end_s": 20.64, "all_asr_last_word_s": 30.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.15217391304347827, "source_identity_reward": 0.19833895395797854, "source_start_sample": 3840, "source_end_sample": 997920, "first_word_output_sample": 3840, "last_word_output_sample": 986880, "samples": 994080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 10.01, "added_silence_s": 0.0, "duration_s": 20.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u00-c02.mp3" }, "de-conversation-05-base17-block-free-u00-c03": { "id": "de-conversation-05-base17-block-free-u00-c03", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 87604, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 19.6, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 45, "coverage": 0.9782608695652174, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 925440, "script_start_s": 0.16, "script_end_s": 19.28, "all_asr_last_word_s": 19.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.021739130434782608, "source_identity_reward": 0.4082483722714093, "source_start_sample": 3840, "source_end_sample": 932640, "first_word_output_sample": 3840, "last_word_output_sample": 921600, "samples": 928800, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.17, "added_silence_s": 0.0, "duration_s": 19.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u00-c03.mp3" }, "de-conversation-05-base17-block-free-u00-c04": { "id": "de-conversation-05-base17-block-free-u00-c04", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 87605, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 46, "matched_tokens": 45, "coverage": 0.9782608695652174, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 929280, "script_start_s": 0.08, "script_end_s": 19.36, "all_asr_last_word_s": 19.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.021739130434782608, "source_identity_reward": 0.2943101106304616, "source_start_sample": 0, "source_end_sample": 936480, "first_word_output_sample": 3840, "last_word_output_sample": 929280, "samples": 936480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 40.49, "added_silence_s": 0.0, "duration_s": 19.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u00-c04.mp3" }, "de-conversation-05-base17-block-free-u00-c05": { "id": "de-conversation-05-base17-block-free-u00-c05", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 87606, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 46, "matched_tokens": 42, "coverage": 0.9130434782608695, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 946560, "script_start_s": 0.0, "script_end_s": 19.72, "all_asr_last_word_s": 48.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.34782608695652173, "source_identity_reward": 0.2804842455614167, "source_start_sample": 0, "source_end_sample": 953760, "first_word_output_sample": 0, "last_word_output_sample": 946560, "samples": 953760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 40.13, "added_silence_s": 0.0, "duration_s": 19.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u00-c05.mp3" }, "de-conversation-05-base17-block-free-u00-c06": { "id": "de-conversation-05-base17-block-free-u00-c06", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 87607, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 43, "coverage": 0.9347826086956522, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1029120, "script_start_s": 0.08, "script_end_s": 21.44, "all_asr_last_word_s": 21.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06521739130434782, "source_identity_reward": 0.2582935586402138, "source_start_sample": 0, "source_end_sample": 1036320, "first_word_output_sample": 3840, "last_word_output_sample": 1029120, "samples": 1036320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.17, "added_silence_s": 0.0, "duration_s": 21.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u00-c06.mp3" }, "de-conversation-05-base17-block-free-u00-c07": { "id": "de-conversation-05-base17-block-free-u00-c07", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 87608, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 46, "matched_tokens": 45, "coverage": 0.9782608695652174, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1073280, "script_start_s": 0.08, "script_end_s": 22.36, "all_asr_last_word_s": 42.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.13043478260869565, "source_identity_reward": 0.3084009080505191, "source_start_sample": 0, "source_end_sample": 1080480, "first_word_output_sample": 3840, "last_word_output_sample": 1073280, "samples": 1080480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 37.49, "added_silence_s": 0.0, "duration_s": 22.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u00-c07.mp3" }, "de-conversation-01-base17-block-free-u02-c00": { "id": "de-conversation-01-base17-block-free-u02-c00", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 83633, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 936960, "script_start_s": 0.16, "script_end_s": 19.52, "all_asr_last_word_s": 24.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.21428571428571427, "source_identity_reward": 0.18829337061102216, "source_start_sample": 3840, "source_end_sample": 944160, "first_word_output_sample": 3840, "last_word_output_sample": 933120, "samples": 940320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.33, "added_silence_s": 0.0, "duration_s": 19.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u02-c00.mp3" }, "de-conversation-01-base17-block-free-u02-c01": { "id": "de-conversation-01-base17-block-free-u02-c01", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 83634, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 54, "coverage": 0.9642857142857143, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 933120, "script_start_s": 0.0, "script_end_s": 19.44, "all_asr_last_word_s": 22.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.16071428571428573, "source_identity_reward": 0.3388550347565423, "source_start_sample": 0, "source_end_sample": 940320, "first_word_output_sample": 0, "last_word_output_sample": 933120, "samples": 940320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 40.41, "added_silence_s": 0.0, "duration_s": 19.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u02-c01.mp3" }, "de-conversation-01-base17-block-free-u02-c02": { "id": "de-conversation-01-base17-block-free-u02-c02", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 83635, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 967680, "script_start_s": 0.0, "script_end_s": 20.16, "all_asr_last_word_s": 20.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3329124934144392, "source_start_sample": 0, "source_end_sample": 974880, "first_word_output_sample": 0, "last_word_output_sample": 967680, "samples": 974880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 39.69, "added_silence_s": 0.0, "duration_s": 20.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u02-c02.mp3" }, "de-conversation-01-base17-block-free-u02-c03": { "id": "de-conversation-01-base17-block-free-u02-c03", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 83636, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 27.84, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 933120, "script_start_s": 0.0, "script_end_s": 19.44, "all_asr_last_word_s": 25.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.23214285714285715, "source_identity_reward": 0.2428530761467842, "source_start_sample": 0, "source_end_sample": 940320, "first_word_output_sample": 0, "last_word_output_sample": 933120, "samples": 940320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 8.25, "added_silence_s": 0.0, "duration_s": 19.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u02-c03.mp3" }, "de-conversation-01-base17-block-free-u02-c04": { "id": "de-conversation-01-base17-block-free-u02-c04", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 83637, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 910080, "script_start_s": 0.0, "script_end_s": 18.96, "all_asr_last_word_s": 23.720000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.16071428571428573, "source_identity_reward": 0.20419391901881817, "source_start_sample": 0, "source_end_sample": 917280, "first_word_output_sample": 0, "last_word_output_sample": 910080, "samples": 917280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 40.89, "added_silence_s": 0.0, "duration_s": 19.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u02-c04.mp3" }, "de-conversation-01-base17-block-free-u02-c05": { "id": "de-conversation-01-base17-block-free-u02-c05", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 83638, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 28.08, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 54, "coverage": 0.9642857142857143, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 956160, "script_start_s": 0.08, "script_end_s": 19.92, "all_asr_last_word_s": 26.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.26785714285714285, "source_identity_reward": 0.20565995648422114, "source_start_sample": 0, "source_end_sample": 963360, "first_word_output_sample": 3840, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 8.01, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u02-c05.mp3" }, "de-conversation-01-base17-block-free-u02-c06": { "id": "de-conversation-01-base17-block-free-u02-c06", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 83639, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 29.12, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 26.080000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.23214285714285715, "source_identity_reward": 0.23185685551659213, "source_start_sample": 0, "source_end_sample": 963360, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 9.05, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u02-c06.mp3" }, "de-conversation-01-base17-block-free-u02-c07": { "id": "de-conversation-01-base17-block-free-u02-c07", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 83640, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 53, "coverage": 0.9464285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 965760, "script_start_s": 0.0, "script_end_s": 20.12, "all_asr_last_word_s": 21.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.14285714285714285, "source_identity_reward": 0.20642251218710705, "source_start_sample": 0, "source_end_sample": 972960, "first_word_output_sample": 0, "last_word_output_sample": 965760, "samples": 972960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 39.73, "added_silence_s": 0.0, "duration_s": 20.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u02-c07.mp3" }, "de-conversation-05-base17-segment-free-u06-c00": { "id": "de-conversation-05-base17-segment-free-u06-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87197, "plain": "Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 384000, "script_start_s": 0.0, "script_end_s": 8.0, "all_asr_last_word_s": 8.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.665158861014477, "source_start_sample": 0, "source_end_sample": 391200, "first_word_output_sample": 0, "last_word_output_sample": 384000, "samples": 391200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 21.85, "added_silence_s": 0.0, "duration_s": 8.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u06-c00.mp3" }, "en-conversation-04-podcast-segment-timed-u07-c00": { "id": "en-conversation-04-podcast-segment-timed-u07-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76213, "plain": "A reassuringly human approach to a very peculiar little machine.", "source_duration_s": 3.52, "source_frame_cap": false, "alignment": { "expected_tokens": 10, "matched_tokens": 10, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 165120, "script_start_s": 0.16, "script_end_s": 3.44, "all_asr_last_word_s": 3.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.15191460205083387, "source_start_sample": 3840, "source_end_sample": 168960, "first_word_output_sample": 3840, "last_word_output_sample": 161280, "samples": 168480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-timed-u07-c00.mp3" }, "de-conversation-02-podcast-segment-timed-u03-c00": { "id": "de-conversation-02-podcast-segment-timed-u03-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84149, "plain": "Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen?", "source_duration_s": 4.48, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 211200, "script_start_s": 0.0, "script_end_s": 4.4, "all_asr_last_word_s": 4.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5143678076328548, "source_start_sample": 0, "source_end_sample": 215040, "first_word_output_sample": 0, "last_word_output_sample": 211200, "samples": 218400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-timed-u03-c00.mp3" }, "de-conversation-02-podcast-segment-free-u04-c00": { "id": "de-conversation-02-podcast-segment-free-u04-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84165, "plain": "Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 8.88, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 291840, "script_start_s": 0.0, "script_end_s": 6.08, "all_asr_last_word_s": 6.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5981231481362789, "source_start_sample": 0, "source_end_sample": 299040, "first_word_output_sample": 0, "last_word_output_sample": 291840, "samples": 299040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.65, "added_silence_s": 0.0, "duration_s": 6.23, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-free-u04-c00.mp3" }, "en-conversation-02-podcast-segment-free-u11-c00": { "id": "en-conversation-02-podcast-segment-free-u11-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 6.96, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 261120, "script_start_s": 0.16, "script_end_s": 5.44, "all_asr_last_word_s": 5.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.20229794302788673, "source_start_sample": 3840, "source_end_sample": 268320, "first_word_output_sample": 3840, "last_word_output_sample": 257280, "samples": 264480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.37, "added_silence_s": 0.0, "duration_s": 5.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-free-u11-c00.mp3" }, "de-conversation-05-podcast-segment-timed-u07-c00": { "id": "de-conversation-05-podcast-segment-timed-u07-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87213, "plain": "Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 234240, "script_start_s": 0.16, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.48507761068217903, "source_start_sample": 3840, "source_end_sample": 241440, "first_word_output_sample": 3840, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-timed-u07-c00.mp3" }, "de-conversation-05-podcast-segment-timed-u06-c00": { "id": "de-conversation-05-podcast-segment-timed-u06-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87197, "plain": "Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind.", "source_duration_s": 4.88, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 230400, "script_start_s": 0.0, "script_end_s": 4.8, "all_asr_last_word_s": 4.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7334083754938365, "source_start_sample": 0, "source_end_sample": 234240, "first_word_output_sample": 0, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-timed-u06-c00.mp3" }, "en-conversation-02-podcast-segment-free-u10-c00": { "id": "en-conversation-02-podcast-segment-free-u10-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 264960, "script_start_s": 0.16, "script_end_s": 5.52, "all_asr_last_word_s": 28.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 2.142857142857143, "source_identity_reward": 0.11829995081707081, "source_start_sample": 3840, "source_end_sample": 272160, "first_word_output_sample": 3840, "last_word_output_sample": 261120, "samples": 268320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 24.33, "added_silence_s": 0.0, "duration_s": 5.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-free-u10-c00.mp3" }, "en-conversation-04-podcast-block-timed-u00-c00": { "id": "en-conversation-04-podcast-block-timed-u00-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 76601, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.64, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 65, "coverage": 0.9558823529411765, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04411764705882353, "source_identity_reward": 0.5229466733857365 }, "en-conversation-04-podcast-block-timed-u00-c01": { "id": "en-conversation-04-podcast-block-timed-u00-c01", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 76602, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.4, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 66, "coverage": 0.9705882352941176, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1071360, "script_start_s": 0.0, "script_end_s": 22.32, "all_asr_last_word_s": 22.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029411764705882353, "source_identity_reward": 0.5044181960976187, "source_start_sample": 0, "source_end_sample": 1075200, "first_word_output_sample": 0, "last_word_output_sample": 1071360, "samples": 1078560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 22.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-timed-u00-c01.mp3" }, "en-conversation-04-podcast-block-timed-u00-c02": { "id": "en-conversation-04-podcast-block-timed-u00-c02", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 76603, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.48, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 65, "coverage": 0.9558823529411765, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04411764705882353, "source_identity_reward": 0.4760832679613026 }, "en-conversation-04-podcast-block-timed-u00-c03": { "id": "en-conversation-04-podcast-block-timed-u00-c03", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 76604, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.56, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 65, "coverage": 0.9558823529411765, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04411764705882353, "source_identity_reward": 0.5298331089482201 }, "en-conversation-04-podcast-block-timed-u00-c04": { "id": "en-conversation-04-podcast-block-timed-u00-c04", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 76605, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.4, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 66, "coverage": 0.9705882352941176, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1071360, "script_start_s": 0.16, "script_end_s": 22.32, "all_asr_last_word_s": 22.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029411764705882353, "source_identity_reward": 0.5686175963185385, "source_start_sample": 3840, "source_end_sample": 1075200, "first_word_output_sample": 3840, "last_word_output_sample": 1067520, "samples": 1074720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 22.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-timed-u00-c04.mp3" }, "en-conversation-04-podcast-block-timed-u00-c05": { "id": "en-conversation-04-podcast-block-timed-u00-c05", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 76606, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.4, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1071360, "script_start_s": 0.0, "script_end_s": 22.32, "all_asr_last_word_s": 22.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.5320907859283609, "source_start_sample": 0, "source_end_sample": 1075200, "first_word_output_sample": 0, "last_word_output_sample": 1071360, "samples": 1078560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 22.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-timed-u00-c05.mp3" }, "en-conversation-04-podcast-block-timed-u00-c06": { "id": "en-conversation-04-podcast-block-timed-u00-c06", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 76607, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.72, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 63, "coverage": 0.9264705882352942, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08823529411764706, "source_identity_reward": 0.47279744760329806 }, "en-conversation-04-podcast-block-timed-u00-c07": { "id": "en-conversation-04-podcast-block-timed-u00-c07", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 76608, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.48, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 65, "coverage": 0.9558823529411765, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04411764705882353, "source_identity_reward": 0.49712266420930734 }, "de-conversation-03-podcast-segment-free-u09-c00": { "id": "de-conversation-03-podcast-segment-free-u09-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 20.56, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 337920, "script_start_s": 0.0, "script_end_s": 7.04, "all_asr_last_word_s": 11.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.35714285714285715, "source_identity_reward": 0.2901884425129731, "source_start_sample": 0, "source_end_sample": 345120, "first_word_output_sample": 0, "last_word_output_sample": 337920, "samples": 345120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 13.37, "added_silence_s": 0.0, "duration_s": 7.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-free-u09-c00.mp3" }, "de-conversation-05-podcast-segment-free-u14-c00": { "id": "de-conversation-05-podcast-segment-free-u14-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87325, "plain": "Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein.", "source_duration_s": 10.08, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 395520, "script_start_s": 0.08, "script_end_s": 8.24, "all_asr_last_word_s": 8.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u14-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7472906743182771, "source_start_sample": 0, "source_end_sample": 402720, "first_word_output_sample": 3840, "last_word_output_sample": 395520, "samples": 402720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.69, "added_silence_s": 0.0, "duration_s": 8.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u14-c00.mp3" }, "en-conversation-02-podcast-block-timed-u02-c00": { "id": "en-conversation-02-podcast-block-timed-u02-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 74633, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 460800, "script_start_s": 0.0, "script_end_s": 9.6, "all_asr_last_word_s": 9.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.44953436227087173, "source_start_sample": 0, "source_end_sample": 464640, "first_word_output_sample": 0, "last_word_output_sample": 460800, "samples": 468000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u02-c00.mp3" }, "en-conversation-02-podcast-block-timed-u02-c01": { "id": "en-conversation-02-podcast-block-timed-u02-c01", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 74634, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 460800, "script_start_s": 0.0, "script_end_s": 9.6, "all_asr_last_word_s": 9.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4775861179122517, "source_start_sample": 0, "source_end_sample": 464640, "first_word_output_sample": 0, "last_word_output_sample": 460800, "samples": 468000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u02-c01.mp3" }, "en-conversation-02-podcast-block-timed-u02-c02": { "id": "en-conversation-02-podcast-block-timed-u02-c02", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 74635, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.6, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 453120, "script_start_s": 0.16, "script_end_s": 9.44, "all_asr_last_word_s": 9.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4741742970811068, "source_start_sample": 3840, "source_end_sample": 460320, "first_word_output_sample": 3840, "last_word_output_sample": 449280, "samples": 456480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u02-c02.mp3" }, "en-conversation-02-podcast-block-timed-u02-c03": { "id": "en-conversation-02-podcast-block-timed-u02-c03", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 74636, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 456960, "script_start_s": 0.0, "script_end_s": 9.52, "all_asr_last_word_s": 9.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5444638672641234, "source_start_sample": 0, "source_end_sample": 464160, "first_word_output_sample": 0, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u02-c03.mp3" }, "en-conversation-02-podcast-block-timed-u02-c04": { "id": "en-conversation-02-podcast-block-timed-u02-c04", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 74637, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 453120, "script_start_s": 0.0, "script_end_s": 9.44, "all_asr_last_word_s": 9.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4183838968811795, "source_start_sample": 0, "source_end_sample": 460320, "first_word_output_sample": 0, "last_word_output_sample": 453120, "samples": 460320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 9.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u02-c04.mp3" }, "en-conversation-02-podcast-block-timed-u02-c05": { "id": "en-conversation-02-podcast-block-timed-u02-c05", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 74638, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 460800, "script_start_s": 0.0, "script_end_s": 9.6, "all_asr_last_word_s": 9.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4263763216025189, "source_start_sample": 0, "source_end_sample": 464640, "first_word_output_sample": 0, "last_word_output_sample": 460800, "samples": 468000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u02-c05.mp3" }, "en-conversation-02-podcast-block-timed-u02-c06": { "id": "en-conversation-02-podcast-block-timed-u02-c06", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 74639, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 453120, "script_start_s": 0.0, "script_end_s": 9.44, "all_asr_last_word_s": 9.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6059917230288768, "source_start_sample": 0, "source_end_sample": 460320, "first_word_output_sample": 0, "last_word_output_sample": 453120, "samples": 460320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 9.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u02-c06.mp3" }, "en-conversation-02-podcast-block-timed-u02-c07": { "id": "en-conversation-02-podcast-block-timed-u02-c07", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 74640, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.6, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 453120, "script_start_s": 0.0, "script_end_s": 9.44, "all_asr_last_word_s": 9.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.43966957255809125, "source_start_sample": 0, "source_end_sample": 460320, "first_word_output_sample": 0, "last_word_output_sample": 453120, "samples": 460320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u02-c07.mp3" }, "de-conversation-02-podcast-segment-free-u08-c00": { "id": "de-conversation-02-podcast-segment-free-u08-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 307200, "script_start_s": 0.16, "script_end_s": 6.4, "all_asr_last_word_s": 6.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6904786602673634, "source_start_sample": 3840, "source_end_sample": 314400, "first_word_output_sample": 3840, "last_word_output_sample": 303360, "samples": 310560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.45, "added_silence_s": 0.0, "duration_s": 6.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-free-u08-c00.mp3" }, "de-conversation-04-podcast-segment-free-u00-c00": { "id": "de-conversation-04-podcast-segment-free-u00-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86101, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 257280, "script_start_s": 0.16, "script_end_s": 5.36, "all_asr_last_word_s": 7.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.25, "source_identity_reward": 0.5786660229510471, "source_start_sample": 3840, "source_end_sample": 264480, "first_word_output_sample": 3840, "last_word_output_sample": 253440, "samples": 260640, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 24.49, "added_silence_s": 0.0, "duration_s": 5.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-free-u00-c00.mp3" }, "en-conversation-03-podcast-segment-timed-u03-c00": { "id": "en-conversation-03-podcast-segment-timed-u03-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75149, "plain": "I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 222720, "script_start_s": 0.16, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.03034380941386743, "source_start_sample": 3840, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-timed-u03-c00.mp3" }, "en-conversation-04-podcast-segment-timed-u02-c00": { "id": "en-conversation-04-podcast-segment-timed-u02-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76133, "plain": "No. The codec advances in eighty-millisecond frames, and requested timing can still miss.", "source_duration_s": 4.48, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 12, "coverage": 0.8571428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 211200, "script_start_s": 0.0, "script_end_s": 4.4, "all_asr_last_word_s": 4.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.14285714285714285, "source_identity_reward": 0.2952670140401952, "source_start_sample": 0, "source_end_sample": 215040, "first_word_output_sample": 0, "last_word_output_sample": 211200, "samples": 218400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-timed-u02-c00.mp3" }, "en-conversation-03-podcast-segment-free-u02-c00": { "id": "en-conversation-03-podcast-segment-free-u02-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75133, "plain": "Yes. The model sees speaker-labelled references and, when available, earlier dialogue.", "source_duration_s": 5.68, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 11, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 268800, "script_start_s": 0.16, "script_end_s": 5.6, "all_asr_last_word_s": 5.6000000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.42822575689859405, "source_start_sample": 3840, "source_end_sample": 272640, "first_word_output_sample": 3840, "last_word_output_sample": 264960, "samples": 272160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-free-u02-c00.mp3" }, "en-conversation-01-podcast-segment-free-u06-c00": { "id": "en-conversation-01-podcast-segment-free-u06-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73197, "plain": "The new training mixes real conversations with acting examples to retain both formats.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 268800, "script_start_s": 0.0, "script_end_s": 5.6, "all_asr_last_word_s": 5.6000000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5555887140618218, "source_start_sample": 0, "source_end_sample": 276000, "first_word_output_sample": 0, "last_word_output_sample": 268800, "samples": 276000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.25, "added_silence_s": 0.0, "duration_s": 5.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-free-u06-c00.mp3" }, "en-conversation-02-podcast-segment-timed-u10-c00": { "id": "en-conversation-02-podcast-segment-timed-u10-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 4.8, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 226560, "script_start_s": 0.16, "script_end_s": 4.72, "all_asr_last_word_s": 4.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3833779024003347, "source_start_sample": 3840, "source_end_sample": 230400, "first_word_output_sample": 3840, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-timed-u10-c00.mp3" }, "de-conversation-02-podcast-segment-free-u02-c00": { "id": "de-conversation-02-podcast-segment-free-u02-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84133, "plain": "Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog.", "source_duration_s": 9.84, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 10, "coverage": 0.8333333333333334, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 311040, "script_start_s": 0.16, "script_end_s": 6.48, "all_asr_last_word_s": 7.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5, "source_identity_reward": 0.41893007183028735, "source_start_sample": 3840, "source_end_sample": 318240, "first_word_output_sample": 3840, "last_word_output_sample": 307200, "samples": 314400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 3.21, "added_silence_s": 0.0, "duration_s": 6.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-free-u02-c00.mp3" }, "en-conversation-04-podcast-segment-timed-u11-c00": { "id": "en-conversation-04-podcast-segment-timed-u11-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 234240, "script_start_s": 0.16, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.06853907737601357, "source_start_sample": 3840, "source_end_sample": 241440, "first_word_output_sample": 3840, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-timed-u11-c00.mp3" }, "de-conversation-01-podcast-block-timed-u00-c00": { "id": "de-conversation-01-podcast-block-timed-u00-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 83601, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.68, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 41, "coverage": 0.9318181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 744960, "script_start_s": 0.0, "script_end_s": 15.52, "all_asr_last_word_s": 15.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06818181818181818, "source_identity_reward": 0.7236180659280173, "source_start_sample": 0, "source_end_sample": 752160, "first_word_output_sample": 0, "last_word_output_sample": 744960, "samples": 752160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 15.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u00-c00.mp3" }, "de-conversation-01-podcast-block-timed-u00-c01": { "id": "de-conversation-01-podcast-block-timed-u00-c01", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 83602, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.68, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 42, "coverage": 0.9545454545454546, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 748800, "script_start_s": 0.0, "script_end_s": 15.6, "all_asr_last_word_s": 15.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.045454545454545456, "source_identity_reward": 0.8041886831139752, "source_start_sample": 0, "source_end_sample": 752640, "first_word_output_sample": 0, "last_word_output_sample": 748800, "samples": 756000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 15.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u00-c01.mp3" }, "de-conversation-01-podcast-block-timed-u00-c02": { "id": "de-conversation-01-podcast-block-timed-u00-c02", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 83603, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.76, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 42, "coverage": 0.9545454545454546, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 748800, "script_start_s": 0.0, "script_end_s": 15.6, "all_asr_last_word_s": 15.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.045454545454545456, "source_identity_reward": 0.7136047589503524, "source_start_sample": 0, "source_end_sample": 756000, "first_word_output_sample": 0, "last_word_output_sample": 748800, "samples": 756000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 15.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u00-c02.mp3" }, "de-conversation-01-podcast-block-timed-u00-c03": { "id": "de-conversation-01-podcast-block-timed-u00-c03", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 83604, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.68, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 41, "coverage": 0.9318181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 744960, "script_start_s": 0.0, "script_end_s": 15.52, "all_asr_last_word_s": 15.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09090909090909091, "source_identity_reward": 0.6939903661512434, "source_start_sample": 0, "source_end_sample": 752160, "first_word_output_sample": 0, "last_word_output_sample": 744960, "samples": 752160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 15.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u00-c03.mp3" }, "de-conversation-01-podcast-block-timed-u00-c04": { "id": "de-conversation-01-podcast-block-timed-u00-c04", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 83605, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.76, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 42, "coverage": 0.9545454545454546, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 748800, "script_start_s": 0.0, "script_end_s": 15.6, "all_asr_last_word_s": 15.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.045454545454545456, "source_identity_reward": 0.8111572122202735, "source_start_sample": 0, "source_end_sample": 756000, "first_word_output_sample": 0, "last_word_output_sample": 748800, "samples": 756000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 15.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u00-c04.mp3" }, "de-conversation-01-podcast-block-timed-u00-c05": { "id": "de-conversation-01-podcast-block-timed-u00-c05", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 83606, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.68, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 41, "coverage": 0.9318181818181818, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09090909090909091, "source_identity_reward": 0.733545504722513 }, "de-conversation-01-podcast-block-timed-u00-c06": { "id": "de-conversation-01-podcast-block-timed-u00-c06", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 83607, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.68, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 41, "coverage": 0.9318181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 748800, "script_start_s": 0.0, "script_end_s": 15.6, "all_asr_last_word_s": 15.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06818181818181818, "source_identity_reward": 0.7771664227549883, "source_start_sample": 0, "source_end_sample": 752640, "first_word_output_sample": 0, "last_word_output_sample": 748800, "samples": 756000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 15.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u00-c06.mp3" }, "de-conversation-01-podcast-block-timed-u00-c07": { "id": "de-conversation-01-podcast-block-timed-u00-c07", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 83608, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.68, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 40, "coverage": 0.9090909090909091, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 748800, "script_start_s": 0.0, "script_end_s": 15.6, "all_asr_last_word_s": 15.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.11363636363636363, "source_identity_reward": 0.7187786768443486, "source_start_sample": 0, "source_end_sample": 752640, "first_word_output_sample": 0, "last_word_output_sample": 748800, "samples": 756000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 15.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u00-c07.mp3" }, "en-conversation-04-podcast-segment-free-u04-c00": { "id": "en-conversation-04-podcast-segment-free-u04-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76165, "plain": "We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 307200, "script_start_s": 0.16, "script_end_s": 6.4, "all_asr_last_word_s": 29.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 3.25, "source_identity_reward": 0.09260319673960395, "source_start_sample": 3840, "source_end_sample": 314400, "first_word_output_sample": 3840, "last_word_output_sample": 303360, "samples": 310560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.45, "added_silence_s": 0.0, "duration_s": 6.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-free-u04-c00.mp3" }, "en-conversation-05-podcast-segment-timed-u11-c00": { "id": "en-conversation-05-podcast-segment-timed-u11-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 238080, "script_start_s": 0.16, "script_end_s": 4.96, "all_asr_last_word_s": 4.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.12267010910868664, "source_start_sample": 3840, "source_end_sample": 241920, "first_word_output_sample": 3840, "last_word_output_sample": 234240, "samples": 241440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-timed-u11-c00.mp3" }, "en-conversation-03-podcast-segment-timed-u11-c00": { "id": "en-conversation-03-podcast-segment-timed-u11-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 238080, "script_start_s": 0.16, "script_end_s": 4.96, "all_asr_last_word_s": 4.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.30477022574605306, "source_start_sample": 3840, "source_end_sample": 241920, "first_word_output_sample": 3840, "last_word_output_sample": 234240, "samples": 241440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-timed-u11-c00.mp3" }, "de-conversation-04-podcast-segment-free-u08-c00": { "id": "de-conversation-04-podcast-segment-free-u08-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 14.08, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 311040, "script_start_s": 0.0, "script_end_s": 6.48, "all_asr_last_word_s": 6.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7153655460517596, "source_start_sample": 0, "source_end_sample": 318240, "first_word_output_sample": 0, "last_word_output_sample": 311040, "samples": 318240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 7.45, "added_silence_s": 0.0, "duration_s": 6.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-free-u08-c00.mp3" }, "en-conversation-05-podcast-segment-timed-u01-c00": { "id": "en-conversation-05-podcast-segment-timed-u01-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77117, "plain": "Same model family, different voices, and a different emotional rhythm in each language?", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 207360, "script_start_s": 0.16, "script_end_s": 4.32, "all_asr_last_word_s": 4.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3462745425881181, "source_start_sample": 3840, "source_end_sample": 211200, "first_word_output_sample": 3840, "last_word_output_sample": 203520, "samples": 210720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-timed-u01-c00.mp3" }, "en-conversation-03-podcast-segment-timed-u06-c00": { "id": "en-conversation-03-podcast-segment-timed-u06-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75197, "plain": "References guide identity; acting directions guide expression. Neither is a magic guarantee.", "source_duration_s": 4.24, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 199680, "script_start_s": 0.0, "script_end_s": 4.16, "all_asr_last_word_s": 4.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3923704732316169, "source_start_sample": 0, "source_end_sample": 203520, "first_word_output_sample": 0, "last_word_output_sample": 199680, "samples": 206880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-timed-u06-c00.mp3" }, "en-conversation-02-podcast-segment-timed-u02-c00": { "id": "en-conversation-02-podcast-segment-timed-u02-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74133, "plain": "Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue.", "source_duration_s": 4.24, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 12, "coverage": 0.9230769230769231, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 199680, "script_start_s": 0.0, "script_end_s": 4.16, "all_asr_last_word_s": 4.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07692307692307693, "source_identity_reward": 0.4837974539010395, "source_start_sample": 0, "source_end_sample": 203520, "first_word_output_sample": 0, "last_word_output_sample": 199680, "samples": 206880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-timed-u02-c00.mp3" }, "en-conversation-02-podcast-segment-free-u06-c00": { "id": "en-conversation-02-podcast-segment-free-u06-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74197, "plain": "That is why the reference clips sit beside the output. You can compare for yourself.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 245760, "script_start_s": 0.24, "script_end_s": 5.12, "all_asr_last_word_s": 5.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1972602027184991, "source_start_sample": 7680, "source_end_sample": 252960, "first_word_output_sample": 3840, "last_word_output_sample": 238080, "samples": 245280, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 24.73, "added_silence_s": 0.0, "duration_s": 5.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-free-u06-c00.mp3" }, "en-conversation-05-podcast-segment-free-u03-c00": { "id": "en-conversation-05-podcast-segment-free-u03-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77149, "plain": "I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 13.6, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 307200, "script_start_s": 0.0, "script_end_s": 6.4, "all_asr_last_word_s": 6.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.17505341332444185, "source_start_sample": 0, "source_end_sample": 314400, "first_word_output_sample": 0, "last_word_output_sample": 307200, "samples": 314400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 7.05, "added_silence_s": 0.0, "duration_s": 6.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-free-u03-c00.mp3" }, "en-conversation-01-podcast-segment-timed-u03-c00": { "id": "en-conversation-01-podcast-segment-timed-u03-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73149, "plain": "Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 207360, "script_start_s": 0.16, "script_end_s": 4.32, "all_asr_last_word_s": 4.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.28255859991238585, "source_start_sample": 3840, "source_end_sample": 214560, "first_word_output_sample": 3840, "last_word_output_sample": 203520, "samples": 210720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.17, "added_silence_s": 0.0, "duration_s": 4.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-timed-u03-c00.mp3" }, "en-conversation-02-podcast-block-timed-u00-c00": { "id": "en-conversation-02-podcast-block-timed-u00-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 74601, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.16, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04477611940298507, "source_identity_reward": 0.5442381397963795 }, "en-conversation-02-podcast-block-timed-u00-c01": { "id": "en-conversation-02-podcast-block-timed-u00-c01", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 74602, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.08, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029850746268656716, "source_identity_reward": 0.5658358975446647 }, "en-conversation-02-podcast-block-timed-u00-c02": { "id": "en-conversation-02-podcast-block-timed-u00-c02", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 74603, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.0, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1052160, "script_start_s": 0.0, "script_end_s": 21.92, "all_asr_last_word_s": 21.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029850746268656716, "source_identity_reward": 0.5884383838644986, "source_start_sample": 0, "source_end_sample": 1056000, "first_word_output_sample": 0, "last_word_output_sample": 1052160, "samples": 1059360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 22.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u00-c02.mp3" }, "en-conversation-02-podcast-block-timed-u00-c03": { "id": "en-conversation-02-podcast-block-timed-u00-c03", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 74604, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.08, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029850746268656716, "source_identity_reward": 0.5583953318834763 }, "en-conversation-02-podcast-block-timed-u00-c04": { "id": "en-conversation-02-podcast-block-timed-u00-c04", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 74605, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.0, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1052160, "script_start_s": 0.0, "script_end_s": 21.92, "all_asr_last_word_s": 21.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.5687679967791294, "source_start_sample": 0, "source_end_sample": 1056000, "first_word_output_sample": 0, "last_word_output_sample": 1052160, "samples": 1059360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 22.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u00-c04.mp3" }, "en-conversation-02-podcast-block-timed-u00-c05": { "id": "en-conversation-02-podcast-block-timed-u00-c05", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 74606, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.0, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1052160, "script_start_s": 0.16, "script_end_s": 21.92, "all_asr_last_word_s": 21.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.5907679108300029, "source_start_sample": 3840, "source_end_sample": 1056000, "first_word_output_sample": 3840, "last_word_output_sample": 1048320, "samples": 1055520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u00-c05.mp3" }, "en-conversation-02-podcast-block-timed-u00-c06": { "id": "en-conversation-02-podcast-block-timed-u00-c06", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 74607, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.0, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1052160, "script_start_s": 0.16, "script_end_s": 21.92, "all_asr_last_word_s": 21.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029850746268656716, "source_identity_reward": 0.5206158980062163, "source_start_sample": 3840, "source_end_sample": 1056000, "first_word_output_sample": 3840, "last_word_output_sample": 1048320, "samples": 1055520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u00-c06.mp3" }, "en-conversation-02-podcast-block-timed-u00-c07": { "id": "en-conversation-02-podcast-block-timed-u00-c07", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 74608, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.16, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04477611940298507, "source_identity_reward": 0.5403098608830919 }, "en-conversation-01-podcast-segment-timed-u06-c00": { "id": "en-conversation-01-podcast-segment-timed-u06-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73197, "plain": "The new training mixes real conversations with acting examples to retain both formats.", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 207360, "script_start_s": 0.0, "script_end_s": 4.32, "all_asr_last_word_s": 4.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.38929834159435384, "source_start_sample": 0, "source_end_sample": 211200, "first_word_output_sample": 0, "last_word_output_sample": 207360, "samples": 214560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-timed-u06-c00.mp3" }, "de-conversation-01-podcast-segment-free-u01-c00": { "id": "de-conversation-01-podcast-segment-free-u01-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83117, "plain": "Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht?", "source_duration_s": 8.32, "source_frame_cap": false, "alignment": { "expected_tokens": 10, "matched_tokens": 10, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 249600, "script_start_s": 0.08, "script_end_s": 5.2, "all_asr_last_word_s": 5.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5157786304889652, "source_start_sample": 0, "source_end_sample": 256800, "first_word_output_sample": 3840, "last_word_output_sample": 249600, "samples": 256800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.97, "added_silence_s": 0.0, "duration_s": 5.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-free-u01-c00.mp3" }, "de-conversation-03-podcast-segment-free-u07-c00": { "id": "de-conversation-03-podcast-segment-free-u07-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85213, "plain": "Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 288000, "script_start_s": 0.16, "script_end_s": 6.0, "all_asr_last_word_s": 6.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.366985218048843, "source_start_sample": 3840, "source_end_sample": 295200, "first_word_output_sample": 3840, "last_word_output_sample": 284160, "samples": 291360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.85, "added_silence_s": 0.0, "duration_s": 6.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-free-u07-c00.mp3" }, "de-conversation-05-podcast-segment-timed-u01-c00": { "id": "de-conversation-05-podcast-segment-timed-u01-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87117, "plain": "Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus?", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 11, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.31977177992167743 }, "de-conversation-05-podcast-segment-free-u07-c00": { "id": "de-conversation-05-podcast-segment-free-u07-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87213, "plain": "Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 6.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 272640, "script_start_s": 0.0, "script_end_s": 5.68, "all_asr_last_word_s": 5.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6011107966296942, "source_start_sample": 0, "source_end_sample": 279840, "first_word_output_sample": 0, "last_word_output_sample": 272640, "samples": 279840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.81, "added_silence_s": 0.0, "duration_s": 5.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u07-c00.mp3" }, "de-conversation-05-podcast-block-free-u00-c00": { "id": "de-conversation-05-podcast-block-free-u00-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 87601, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 20.8, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 44, "coverage": 0.9565217391304348, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 990720, "script_start_s": 0.0, "script_end_s": 20.64, "all_asr_last_word_s": 20.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.043478260869565216, "source_identity_reward": 0.77698234734012, "source_start_sample": 0, "source_end_sample": 997920, "first_word_output_sample": 0, "last_word_output_sample": 990720, "samples": 997920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u00-c00.mp3" }, "de-conversation-05-podcast-block-free-u00-c01": { "id": "de-conversation-05-podcast-block-free-u00-c01", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 87602, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 22.48, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 46, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1017600, "script_start_s": 0.0, "script_end_s": 21.2, "all_asr_last_word_s": 22.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.021739130434782608, "source_identity_reward": 0.5949699622767859, "source_start_sample": 0, "source_end_sample": 1024800, "first_word_output_sample": 0, "last_word_output_sample": 1017600, "samples": 1024800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.13, "added_silence_s": 0.0, "duration_s": 21.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u00-c01.mp3" }, "de-conversation-05-podcast-block-free-u00-c02": { "id": "de-conversation-05-podcast-block-free-u00-c02", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 87603, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 33.84, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 45, "coverage": 0.9782608695652174, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 990720, "script_start_s": 0.0, "script_end_s": 20.64, "all_asr_last_word_s": 31.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.15217391304347827, "source_identity_reward": 0.6878005252003196, "source_start_sample": 0, "source_end_sample": 997920, "first_word_output_sample": 0, "last_word_output_sample": 990720, "samples": 997920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 13.05, "added_silence_s": 0.0, "duration_s": 20.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u00-c02.mp3" }, "de-conversation-05-podcast-block-free-u00-c03": { "id": "de-conversation-05-podcast-block-free-u00-c03", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 87604, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 20.8, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 43, "coverage": 0.9347826086956522, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 990720, "script_start_s": 0.0, "script_end_s": 20.64, "all_asr_last_word_s": 20.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06521739130434782, "source_identity_reward": 0.7493991925365222, "source_start_sample": 0, "source_end_sample": 997920, "first_word_output_sample": 0, "last_word_output_sample": 990720, "samples": 997920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u00-c03.mp3" }, "de-conversation-05-podcast-block-free-u00-c04": { "id": "de-conversation-05-podcast-block-free-u00-c04", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 87605, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 20.8, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 45, "coverage": 0.9782608695652174, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 990720, "script_start_s": 0.0, "script_end_s": 20.64, "all_asr_last_word_s": 20.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.021739130434782608, "source_identity_reward": 0.7692190435898305, "source_start_sample": 0, "source_end_sample": 997920, "first_word_output_sample": 0, "last_word_output_sample": 990720, "samples": 997920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u00-c04.mp3" }, "de-conversation-05-podcast-block-free-u00-c05": { "id": "de-conversation-05-podcast-block-free-u00-c05", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 87606, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 34.72, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 44, "coverage": 0.9565217391304348, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 960000, "script_start_s": 0.0, "script_end_s": 20.0, "all_asr_last_word_s": 33.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2826086956521739, "source_identity_reward": 0.5594259868201594, "source_start_sample": 0, "source_end_sample": 967200, "first_word_output_sample": 0, "last_word_output_sample": 960000, "samples": 967200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 14.57, "added_silence_s": 0.0, "duration_s": 20.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u00-c05.mp3" }, "de-conversation-05-podcast-block-free-u00-c06": { "id": "de-conversation-05-podcast-block-free-u00-c06", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 87607, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 21.2, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 43, "coverage": 0.9347826086956522, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 990720, "script_start_s": 0.0, "script_end_s": 20.64, "all_asr_last_word_s": 20.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06521739130434782, "source_identity_reward": 0.7936085506255212, "source_start_sample": 0, "source_end_sample": 997920, "first_word_output_sample": 0, "last_word_output_sample": 990720, "samples": 997920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.41, "added_silence_s": 0.0, "duration_s": 20.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u00-c06.mp3" }, "de-conversation-05-podcast-block-free-u00-c07": { "id": "de-conversation-05-podcast-block-free-u00-c07", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 87608, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 46, "matched_tokens": 43, "coverage": 0.9347826086956522, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 977280, "script_start_s": 0.0, "script_end_s": 20.36, "all_asr_last_word_s": 20.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06521739130434782, "source_identity_reward": 0.7715774036581535, "source_start_sample": 0, "source_end_sample": 984480, "first_word_output_sample": 0, "last_word_output_sample": 977280, "samples": 984480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 39.49, "added_silence_s": 0.0, "duration_s": 20.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u00-c07.mp3" }, "en-conversation-04-podcast-segment-timed-u08-c00": { "id": "en-conversation-04-podcast-segment-timed-u08-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 222720, "script_start_s": 0.16, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.18869150135171586, "source_start_sample": 3840, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-timed-u08-c00.mp3" }, "en-conversation-03-podcast-segment-free-u11-c00": { "id": "en-conversation-03-podcast-segment-free-u11-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 21.44, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 257280, "script_start_s": 0.16, "script_end_s": 5.36, "all_asr_last_word_s": 18.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4666666666666667, "source_identity_reward": 0.21894700045465615, "source_start_sample": 3840, "source_end_sample": 264480, "first_word_output_sample": 3840, "last_word_output_sample": 253440, "samples": 260640, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 15.93, "added_silence_s": 0.0, "duration_s": 5.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-free-u11-c00.mp3" }, "en-conversation-01-podcast-segment-timed-u05-c00": { "id": "en-conversation-01-podcast-segment-timed-u05-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73181, "plain": "That distinction makes me trust the demonstration a little more. Annoyingly sensible of you.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 218880, "script_start_s": 0.16, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.22373433972852225, "source_start_sample": 3840, "source_end_sample": 226080, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-timed-u05-c00.mp3" }, "de-conversation-05-podcast-segment-free-u08-c00": { "id": "de-conversation-05-podcast-segment-free-u08-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 10.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 311040, "script_start_s": 0.16, "script_end_s": 6.48, "all_asr_last_word_s": 10.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.38461538461538464, "source_identity_reward": 0.520575215647668, "source_start_sample": 3840, "source_end_sample": 318240, "first_word_output_sample": 3840, "last_word_output_sample": 307200, "samples": 314400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 4.01, "added_silence_s": 0.0, "duration_s": 6.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u08-c00.mp3" }, "en-conversation-01-podcast-segment-free-u09-c00": { "id": "en-conversation-01-podcast-segment-free-u09-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.64, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 295680, "script_start_s": 0.0, "script_end_s": 6.16, "all_asr_last_word_s": 11.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0, "source_identity_reward": 0.09318082945200709, "source_start_sample": 0, "source_end_sample": 302880, "first_word_output_sample": 0, "last_word_output_sample": 295680, "samples": 302880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 14.33, "added_silence_s": 0.0, "duration_s": 6.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-free-u09-c00.mp3" }, "de-conversation-03-podcast-segment-timed-u09-c00": { "id": "de-conversation-03-podcast-segment-timed-u09-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 234240, "script_start_s": 0.0, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.41987025500722486, "source_start_sample": 0, "source_end_sample": 241440, "first_word_output_sample": 0, "last_word_output_sample": 234240, "samples": 241440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-timed-u09-c00.mp3" }, "en-conversation-05-podcast-segment-timed-u05-c00": { "id": "en-conversation-05-podcast-segment-timed-u05-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77181, "plain": "I want surprise, warmth, a little teasing, and no one reading the stage directions aloud.", "source_duration_s": 5.12, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 241920, "script_start_s": 0.0, "script_end_s": 5.04, "all_asr_last_word_s": 5.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.18384096548821377, "source_start_sample": 0, "source_end_sample": 245760, "first_word_output_sample": 0, "last_word_output_sample": 241920, "samples": 249120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-timed-u05-c00.mp3" }, "en-conversation-04-podcast-segment-timed-u04-c00": { "id": "en-conversation-04-podcast-segment-timed-u04-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76165, "plain": "We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 4.16, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 195840, "script_start_s": 0.0, "script_end_s": 4.08, "all_asr_last_word_s": 4.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4089547701652999, "source_start_sample": 0, "source_end_sample": 199680, "first_word_output_sample": 0, "last_word_output_sample": 195840, "samples": 203040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.23, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-timed-u04-c00.mp3" }, "de-conversation-04-podcast-segment-timed-u07-c00": { "id": "de-conversation-04-podcast-segment-timed-u07-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86213, "plain": "Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine.", "source_duration_s": 3.76, "source_frame_cap": false, "alignment": { "expected_tokens": 10, "matched_tokens": 10, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 176640, "script_start_s": 0.16, "script_end_s": 3.68, "all_asr_last_word_s": 3.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2003747082232602, "source_start_sample": 3840, "source_end_sample": 180480, "first_word_output_sample": 3840, "last_word_output_sample": 172800, "samples": 180000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-timed-u07-c00.mp3" }, "de-conversation-04-podcast-segment-timed-u03-c00": { "id": "de-conversation-04-podcast-segment-timed-u03-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86149, "plain": "Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen.", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 207360, "script_start_s": 0.0, "script_end_s": 4.32, "all_asr_last_word_s": 4.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4770832876987336, "source_start_sample": 0, "source_end_sample": 211200, "first_word_output_sample": 0, "last_word_output_sample": 207360, "samples": 214560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-timed-u03-c00.mp3" }, "de-conversation-03-podcast-segment-timed-u04-c00": { "id": "de-conversation-03-podcast-segment-timed-u04-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85165, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 12, "coverage": 0.9230769230769231, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07692307692307693, "source_identity_reward": 0.664966418997365, "source_start_sample": 0, "source_end_sample": 226080, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-timed-u04-c00.mp3" }, "de-conversation-02-podcast-segment-free-u10-c00": { "id": "de-conversation-02-podcast-segment-free-u10-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 8.4, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 391680, "script_start_s": 0.0, "script_end_s": 8.16, "all_asr_last_word_s": 8.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7772122238850052, "source_start_sample": 0, "source_end_sample": 398880, "first_word_output_sample": 0, "last_word_output_sample": 391680, "samples": 398880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 8.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-free-u10-c00.mp3" }, "de-conversation-02-podcast-block-timed-u00-c00": { "id": "de-conversation-02-podcast-block-timed-u00-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 84601, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.6, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 56, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.7667313406390767 }, "de-conversation-02-podcast-block-timed-u00-c01": { "id": "de-conversation-02-podcast-block-timed-u00-c01", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 84602, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.52, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 58, "coverage": 0.9666666666666667, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1029120, "script_start_s": 0.0, "script_end_s": 21.44, "all_asr_last_word_s": 21.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05, "source_identity_reward": 0.8028703214880547, "source_start_sample": 0, "source_end_sample": 1032960, "first_word_output_sample": 0, "last_word_output_sample": 1029120, "samples": 1036320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-timed-u00-c01.mp3" }, "de-conversation-02-podcast-block-timed-u00-c02": { "id": "de-conversation-02-podcast-block-timed-u00-c02", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 84603, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.6, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 55, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1, "source_identity_reward": 0.7457965795851743 }, "de-conversation-02-podcast-block-timed-u00-c03": { "id": "de-conversation-02-podcast-block-timed-u00-c03", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 84604, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.68, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 53, "coverage": 0.8833333333333333, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.13333333333333333, "source_identity_reward": 0.7138968502548501 }, "de-conversation-02-podcast-block-timed-u00-c04": { "id": "de-conversation-02-podcast-block-timed-u00-c04", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 84605, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.6, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 58, "coverage": 0.9666666666666667, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1032960, "script_start_s": 0.0, "script_end_s": 21.52, "all_asr_last_word_s": 21.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05, "source_identity_reward": 0.7707824514706513, "source_start_sample": 0, "source_end_sample": 1036800, "first_word_output_sample": 0, "last_word_output_sample": 1032960, "samples": 1040160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-timed-u00-c04.mp3" }, "de-conversation-02-podcast-block-timed-u00-c05": { "id": "de-conversation-02-podcast-block-timed-u00-c05", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 84606, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.92, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 57, "coverage": 0.95, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1048320, "script_start_s": 0.0, "script_end_s": 21.84, "all_asr_last_word_s": 21.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.7443670842698596, "source_start_sample": 0, "source_end_sample": 1052160, "first_word_output_sample": 0, "last_word_output_sample": 1048320, "samples": 1055520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-timed-u00-c05.mp3" }, "de-conversation-02-podcast-block-timed-u00-c06": { "id": "de-conversation-02-podcast-block-timed-u00-c06", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 84607, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.68, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 55, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1, "source_identity_reward": 0.7352456673095152 }, "de-conversation-02-podcast-block-timed-u00-c07": { "id": "de-conversation-02-podcast-block-timed-u00-c07", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 84608, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.6, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 55, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.11666666666666667, "source_identity_reward": 0.7430662097250789 }, "en-conversation-04-podcast-segment-timed-u01-c00": { "id": "en-conversation-04-podcast-segment-timed-u01-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76117, "plain": "You are kidding. Do you mean it lands on an exact millisecond?", "source_duration_s": 4.0, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 188160, "script_start_s": 0.16, "script_end_s": 3.92, "all_asr_last_word_s": 3.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.33596742546486535, "source_start_sample": 3840, "source_end_sample": 192000, "first_word_output_sample": 3840, "last_word_output_sample": 184320, "samples": 191520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-timed-u01-c00.mp3" }, "de-conversation-01-podcast-segment-timed-u01-c00": { "id": "de-conversation-01-podcast-segment-timed-u01-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83117, "plain": "Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht?", "source_duration_s": 3.76, "source_frame_cap": false, "alignment": { "expected_tokens": 10, "matched_tokens": 10, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 176640, "script_start_s": 0.0, "script_end_s": 3.68, "all_asr_last_word_s": 3.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6017426687334955, "source_start_sample": 0, "source_end_sample": 180480, "first_word_output_sample": 0, "last_word_output_sample": 176640, "samples": 183840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-timed-u01-c00.mp3" }, "en-conversation-04-podcast-segment-timed-u09-c00": { "id": "en-conversation-04-podcast-segment-timed-u09-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 192000, "script_start_s": 0.16, "script_end_s": 4.0, "all_asr_last_word_s": 4.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.25240241586771683, "source_start_sample": 3840, "source_end_sample": 195840, "first_word_output_sample": 3840, "last_word_output_sample": 188160, "samples": 195360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-timed-u09-c00.mp3" }, "de-conversation-02-podcast-segment-timed-u06-c00": { "id": "de-conversation-02-podcast-segment-timed-u06-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84197, "plain": "Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen.", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 207360, "script_start_s": 0.0, "script_end_s": 4.32, "all_asr_last_word_s": 4.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.738011239795415, "source_start_sample": 0, "source_end_sample": 211200, "first_word_output_sample": 0, "last_word_output_sample": 207360, "samples": 214560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-timed-u06-c00.mp3" }, "en-conversation-04-podcast-block-free-u02-c00": { "id": "en-conversation-04-podcast-block-free-u02-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 76633, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 491520, "script_start_s": 0.0, "script_end_s": 10.24, "all_asr_last_word_s": 17.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2413793103448276, "source_identity_reward": 0.38419230755134703, "source_start_sample": 0, "source_end_sample": 498720, "first_word_output_sample": 0, "last_word_output_sample": 491520, "samples": 498720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 49.61, "added_silence_s": 0.0, "duration_s": 10.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u02-c00.mp3" }, "en-conversation-04-podcast-block-free-u02-c01": { "id": "en-conversation-04-podcast-block-free-u02-c01", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 76634, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 495360, "script_start_s": 0.16, "script_end_s": 10.32, "all_asr_last_word_s": 59.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.9655172413793104, "source_identity_reward": 0.2306104058231995, "source_start_sample": 3840, "source_end_sample": 502560, "first_word_output_sample": 3840, "last_word_output_sample": 491520, "samples": 498720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 49.53, "added_silence_s": 0.0, "duration_s": 10.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u02-c01.mp3" }, "en-conversation-04-podcast-block-free-u02-c02": { "id": "en-conversation-04-podcast-block-free-u02-c02", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 76635, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 11.2, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 472320, "script_start_s": 0.0, "script_end_s": 9.84, "all_asr_last_word_s": 9.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5768200107017174, "source_start_sample": 0, "source_end_sample": 479520, "first_word_output_sample": 0, "last_word_output_sample": 472320, "samples": 479520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.21, "added_silence_s": 0.0, "duration_s": 9.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u02-c02.mp3" }, "en-conversation-04-podcast-block-free-u02-c03": { "id": "en-conversation-04-podcast-block-free-u02-c03", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 76636, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 437760, "script_start_s": 0.16, "script_end_s": 9.12, "all_asr_last_word_s": 59.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 2.6551724137931036, "source_identity_reward": 0.13737042483035844, "source_start_sample": 3840, "source_end_sample": 444960, "first_word_output_sample": 3840, "last_word_output_sample": 433920, "samples": 441120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 50.73, "added_silence_s": 0.0, "duration_s": 9.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u02-c03.mp3" }, "en-conversation-04-podcast-block-free-u02-c04": { "id": "en-conversation-04-podcast-block-free-u02-c04", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 76637, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 476160, "script_start_s": 0.16, "script_end_s": 9.92, "all_asr_last_word_s": 9.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6381277254439324, "source_start_sample": 3840, "source_end_sample": 483360, "first_word_output_sample": 3840, "last_word_output_sample": 472320, "samples": 479520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.25, "added_silence_s": 0.0, "duration_s": 9.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u02-c04.mp3" }, "en-conversation-04-podcast-block-free-u02-c05": { "id": "en-conversation-04-podcast-block-free-u02-c05", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 76638, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 17.52, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 28, "coverage": 0.9655172413793104, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 495360, "script_start_s": 0.08, "script_end_s": 10.32, "all_asr_last_word_s": 10.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.034482758620689655, "source_identity_reward": 0.5757332035497015, "source_start_sample": 0, "source_end_sample": 502560, "first_word_output_sample": 3840, "last_word_output_sample": 495360, "samples": 502560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 7.05, "added_silence_s": 0.0, "duration_s": 10.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u02-c05.mp3" }, "en-conversation-04-podcast-block-free-u02-c06": { "id": "en-conversation-04-podcast-block-free-u02-c06", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 76639, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 14.24, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 460800, "script_start_s": 0.0, "script_end_s": 9.6, "all_asr_last_word_s": 12.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3103448275862069, "source_identity_reward": 0.3583582752762105, "source_start_sample": 0, "source_end_sample": 468000, "first_word_output_sample": 0, "last_word_output_sample": 460800, "samples": 468000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.49, "added_silence_s": 0.0, "duration_s": 9.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u02-c06.mp3" }, "en-conversation-04-podcast-block-free-u02-c07": { "id": "en-conversation-04-podcast-block-free-u02-c07", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 76640, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 10.4, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 476160, "script_start_s": 0.0, "script_end_s": 9.92, "all_asr_last_word_s": 9.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7062984857337453, "source_start_sample": 0, "source_end_sample": 483360, "first_word_output_sample": 0, "last_word_output_sample": 476160, "samples": 483360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.33, "added_silence_s": 0.0, "duration_s": 10.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u02-c07.mp3" }, "de-conversation-05-podcast-segment-free-u04-c00": { "id": "de-conversation-05-podcast-segment-free-u04-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87165, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war.", "source_duration_s": 27.12, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 13, "coverage": 0.9285714285714286, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 387840, "script_start_s": 0.24, "script_end_s": 8.08, "all_asr_last_word_s": 24.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.2142857142857142, "source_identity_reward": 0.27120167904400033, "source_start_sample": 7680, "source_end_sample": 395040, "first_word_output_sample": 3840, "last_word_output_sample": 380160, "samples": 387360, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 18.89, "added_silence_s": 0.0, "duration_s": 8.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u04-c00.mp3" }, "en-conversation-01-podcast-block-timed-u00-c00": { "id": "en-conversation-01-podcast-block-timed-u00-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 73601, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.64, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 49, "coverage": 0.9423076923076923, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 787200, "script_start_s": 0.0, "script_end_s": 16.4, "all_asr_last_word_s": 16.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.057692307692307696, "source_identity_reward": 0.3650621559096169, "source_start_sample": 0, "source_end_sample": 794400, "first_word_output_sample": 0, "last_word_output_sample": 787200, "samples": 794400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 16.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u00-c00.mp3" }, "en-conversation-01-podcast-block-timed-u00-c01": { "id": "en-conversation-01-podcast-block-timed-u00-c01", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 73602, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.56, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 49, "coverage": 0.9423076923076923, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 787200, "script_start_s": 0.0, "script_end_s": 16.4, "all_asr_last_word_s": 16.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.057692307692307696, "source_identity_reward": 0.5364772689179625, "source_start_sample": 0, "source_end_sample": 794400, "first_word_output_sample": 0, "last_word_output_sample": 787200, "samples": 794400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u00-c01.mp3" }, "en-conversation-01-podcast-block-timed-u00-c02": { "id": "en-conversation-01-podcast-block-timed-u00-c02", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 73603, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.56, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 783360, "script_start_s": 0.0, "script_end_s": 16.32, "all_asr_last_word_s": 16.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.057692307692307696, "source_identity_reward": 0.40670020564638343, "source_start_sample": 0, "source_end_sample": 790560, "first_word_output_sample": 0, "last_word_output_sample": 783360, "samples": 790560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 16.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u00-c02.mp3" }, "en-conversation-01-podcast-block-timed-u00-c03": { "id": "en-conversation-01-podcast-block-timed-u00-c03", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 73604, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.56, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 787200, "script_start_s": 0.0, "script_end_s": 16.4, "all_asr_last_word_s": 16.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.5142903715571355, "source_start_sample": 0, "source_end_sample": 794400, "first_word_output_sample": 0, "last_word_output_sample": 787200, "samples": 794400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u00-c03.mp3" }, "en-conversation-01-podcast-block-timed-u00-c04": { "id": "en-conversation-01-podcast-block-timed-u00-c04", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 73605, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.64, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 49, "coverage": 0.9423076923076923, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 791040, "script_start_s": 0.0, "script_end_s": 16.48, "all_asr_last_word_s": 16.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.057692307692307696, "source_identity_reward": 0.4571681581091522, "source_start_sample": 0, "source_end_sample": 798240, "first_word_output_sample": 0, "last_word_output_sample": 791040, "samples": 798240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u00-c04.mp3" }, "en-conversation-01-podcast-block-timed-u00-c05": { "id": "en-conversation-01-podcast-block-timed-u00-c05", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 73606, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.64, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 791040, "script_start_s": 0.0, "script_end_s": 16.48, "all_asr_last_word_s": 16.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.4516292140125006, "source_start_sample": 0, "source_end_sample": 798240, "first_word_output_sample": 0, "last_word_output_sample": 791040, "samples": 798240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u00-c05.mp3" }, "en-conversation-01-podcast-block-timed-u00-c06": { "id": "en-conversation-01-podcast-block-timed-u00-c06", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 73607, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.64, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 49, "coverage": 0.9423076923076923, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 791040, "script_start_s": 0.0, "script_end_s": 16.48, "all_asr_last_word_s": 16.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.057692307692307696, "source_identity_reward": 0.52232213208842, "source_start_sample": 0, "source_end_sample": 798240, "first_word_output_sample": 0, "last_word_output_sample": 791040, "samples": 798240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u00-c06.mp3" }, "en-conversation-01-podcast-block-timed-u00-c07": { "id": "en-conversation-01-podcast-block-timed-u00-c07", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 73608, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.64, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 49, "coverage": 0.9423076923076923, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 791040, "script_start_s": 0.0, "script_end_s": 16.48, "all_asr_last_word_s": 16.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.057692307692307696, "source_identity_reward": 0.48254173856577554, "source_start_sample": 0, "source_end_sample": 798240, "first_word_output_sample": 0, "last_word_output_sample": 791040, "samples": 798240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u00-c07.mp3" }, "de-conversation-05-podcast-segment-timed-u14-c00": { "id": "de-conversation-05-podcast-segment-timed-u14-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87325, "plain": "Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein.", "source_duration_s": 5.76, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 15, "coverage": 0.9375, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 272640, "script_start_s": 0.08, "script_end_s": 5.68, "all_asr_last_word_s": 5.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u14-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1875, "source_identity_reward": 0.6197672044483966, "source_start_sample": 0, "source_end_sample": 276480, "first_word_output_sample": 3840, "last_word_output_sample": 272640, "samples": 279840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-timed-u14-c00.mp3" }, "en-conversation-05-podcast-block-free-u02-c00": { "id": "en-conversation-05-podcast-block-free-u02-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 77633, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 19.84, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 940800, "script_start_s": 0.16, "script_end_s": 19.6, "all_asr_last_word_s": 19.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.47040025453488205, "source_start_sample": 3840, "source_end_sample": 948000, "first_word_output_sample": 3840, "last_word_output_sample": 936960, "samples": 944160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 19.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u02-c00.mp3" }, "en-conversation-05-podcast-block-free-u02-c01": { "id": "en-conversation-05-podcast-block-free-u02-c01", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 77634, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 906240, "script_start_s": 0.0, "script_end_s": 18.88, "all_asr_last_word_s": 58.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.6, "source_identity_reward": 0.3085548879190227, "source_start_sample": 0, "source_end_sample": 913440, "first_word_output_sample": 0, "last_word_output_sample": 906240, "samples": 913440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 40.97, "added_silence_s": 0.0, "duration_s": 19.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u02-c01.mp3" }, "en-conversation-05-podcast-block-free-u02-c02": { "id": "en-conversation-05-podcast-block-free-u02-c02", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 77635, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 23.28, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 864000, "script_start_s": 0.0, "script_end_s": 18.0, "all_asr_last_word_s": 23.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2909090909090909, "source_identity_reward": 0.34461996927761324, "source_start_sample": 0, "source_end_sample": 871200, "first_word_output_sample": 0, "last_word_output_sample": 864000, "samples": 871200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 5.13, "added_silence_s": 0.0, "duration_s": 18.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u02-c02.mp3" }, "en-conversation-05-podcast-block-free-u02-c03": { "id": "en-conversation-05-podcast-block-free-u02-c03", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 77636, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.40107029665328553, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u02-c03.mp3" }, "en-conversation-05-podcast-block-free-u02-c04": { "id": "en-conversation-05-podcast-block-free-u02-c04", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 77637, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.56, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 883200, "script_start_s": 0.16, "script_end_s": 18.4, "all_asr_last_word_s": 18.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.32784606061397725, "source_start_sample": 3840, "source_end_sample": 890400, "first_word_output_sample": 3840, "last_word_output_sample": 879360, "samples": 886560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 18.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u02-c04.mp3" }, "en-conversation-05-podcast-block-free-u02-c05": { "id": "en-conversation-05-podcast-block-free-u02-c05", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 77638, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.96, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 902400, "script_start_s": 0.16, "script_end_s": 18.8, "all_asr_last_word_s": 18.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4433891783912524, "source_start_sample": 3840, "source_end_sample": 909600, "first_word_output_sample": 3840, "last_word_output_sample": 898560, "samples": 905760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 18.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u02-c05.mp3" }, "en-conversation-05-podcast-block-free-u02-c06": { "id": "en-conversation-05-podcast-block-free-u02-c06", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 77639, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 22.48, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1006080, "script_start_s": 0.16, "script_end_s": 20.96, "all_asr_last_word_s": 21.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.3379860085101167, "source_start_sample": 3840, "source_end_sample": 1013280, "first_word_output_sample": 3840, "last_word_output_sample": 1002240, "samples": 1009440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.37, "added_silence_s": 0.0, "duration_s": 21.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u02-c06.mp3" }, "en-conversation-05-podcast-block-free-u02-c07": { "id": "en-conversation-05-podcast-block-free-u02-c07", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 77640, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 20.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 848640, "script_start_s": 0.16, "script_end_s": 17.68, "all_asr_last_word_s": 17.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4576444154421045, "source_start_sample": 3840, "source_end_sample": 855840, "first_word_output_sample": 3840, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 2.49, "added_silence_s": 0.0, "duration_s": 17.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u02-c07.mp3" }, "en-conversation-04-podcast-segment-free-u06-c00": { "id": "en-conversation-04-podcast-segment-free-u06-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76197, "plain": "That is the idea: control the intention, measure the result, and listen carefully.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 264960, "script_start_s": 0.0, "script_end_s": 5.52, "all_asr_last_word_s": 5.5200000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.42350983424149485, "source_start_sample": 0, "source_end_sample": 272160, "first_word_output_sample": 0, "last_word_output_sample": 264960, "samples": 272160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.33, "added_silence_s": 0.0, "duration_s": 5.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-free-u06-c00.mp3" }, "en-conversation-03-podcast-segment-timed-u00-c00": { "id": "en-conversation-03-podcast-segment-timed-u00-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75101, "plain": "Here is the part that interests me: using a short recording to establish each voice.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 234240, "script_start_s": 0.0, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06666666666666667, "source_identity_reward": 0.3201545430926287, "source_start_sample": 0, "source_end_sample": 241440, "first_word_output_sample": 0, "last_word_output_sample": 234240, "samples": 241440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-timed-u00-c00.mp3" }, "de-conversation-04-podcast-segment-timed-u08-c00": { "id": "de-conversation-04-podcast-segment-timed-u08-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 4.8, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 226560, "script_start_s": 0.0, "script_end_s": 4.72, "all_asr_last_word_s": 4.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5343081645182794, "source_start_sample": 0, "source_end_sample": 230400, "first_word_output_sample": 0, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-timed-u08-c00.mp3" }, "de-conversation-03-podcast-segment-free-u06-c00": { "id": "de-conversation-03-podcast-segment-free-u06-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85197, "plain": "Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie.", "source_duration_s": 13.36, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 311040, "script_start_s": 0.08, "script_end_s": 6.48, "all_asr_last_word_s": 6.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.686542865547485, "source_start_sample": 0, "source_end_sample": 318240, "first_word_output_sample": 3840, "last_word_output_sample": 311040, "samples": 318240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 6.73, "added_silence_s": 0.0, "duration_s": 6.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-free-u06-c00.mp3" }, "de-conversation-01-podcast-segment-free-u07-c00": { "id": "de-conversation-01-podcast-segment-free-u07-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83213, "plain": "Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 12, "coverage": 0.8571428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 272640, "script_start_s": 0.24, "script_end_s": 5.68, "all_asr_last_word_s": 5.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2857142857142857, "source_identity_reward": 0.4417922174908287, "source_start_sample": 7680, "source_end_sample": 279840, "first_word_output_sample": 3840, "last_word_output_sample": 264960, "samples": 272160, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 24.17, "added_silence_s": 0.0, "duration_s": 5.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-free-u07-c00.mp3" }, "en-conversation-04-podcast-segment-timed-u10-c00": { "id": "en-conversation-04-podcast-segment-timed-u10-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 222720, "script_start_s": 0.16, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.44836039643994685, "source_start_sample": 3840, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-timed-u10-c00.mp3" }, "de-conversation-03-podcast-block-timed-u02-c00": { "id": "de-conversation-03-podcast-block-timed-u02-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 85633, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8065829053864365, "source_start_sample": 0, "source_end_sample": 960000, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u02-c00.mp3" }, "de-conversation-03-podcast-block-timed-u02-c01": { "id": "de-conversation-03-podcast-block-timed-u02-c01", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 85634, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7910822421944259, "source_start_sample": 0, "source_end_sample": 960000, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u02-c01.mp3" }, "de-conversation-03-podcast-block-timed-u02-c02": { "id": "de-conversation-03-podcast-block-timed-u02-c02", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 85635, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.08, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7710882043860673, "source_start_sample": 0, "source_end_sample": 963360, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u02-c02.mp3" }, "de-conversation-03-podcast-block-timed-u02-c03": { "id": "de-conversation-03-podcast-block-timed-u02-c03", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 85636, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.08, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 960000, "script_start_s": 0.0, "script_end_s": 20.0, "all_asr_last_word_s": 20.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8069117596300328, "source_start_sample": 0, "source_end_sample": 963840, "first_word_output_sample": 0, "last_word_output_sample": 960000, "samples": 967200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u02-c03.mp3" }, "de-conversation-03-podcast-block-timed-u02-c04": { "id": "de-conversation-03-podcast-block-timed-u02-c04", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 85637, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 952320, "script_start_s": 0.0, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7721514179103084, "source_start_sample": 0, "source_end_sample": 959520, "first_word_output_sample": 0, "last_word_output_sample": 952320, "samples": 959520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u02-c04.mp3" }, "de-conversation-03-podcast-block-timed-u02-c05": { "id": "de-conversation-03-podcast-block-timed-u02-c05", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 85638, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.08, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7700768385193417, "source_start_sample": 0, "source_end_sample": 963360, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u02-c05.mp3" }, "de-conversation-03-podcast-block-timed-u02-c06": { "id": "de-conversation-03-podcast-block-timed-u02-c06", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 85639, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7828377766777643, "source_start_sample": 0, "source_end_sample": 960000, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u02-c06.mp3" }, "de-conversation-03-podcast-block-timed-u02-c07": { "id": "de-conversation-03-podcast-block-timed-u02-c07", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 85640, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.08, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7459037428296852, "source_start_sample": 0, "source_end_sample": 963360, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u02-c07.mp3" }, "de-conversation-05-podcast-segment-free-u00-c00": { "id": "de-conversation-05-podcast-segment-free-u00-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87101, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch.", "source_duration_s": 7.04, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 330240, "script_start_s": 0.24, "script_end_s": 6.88, "all_asr_last_word_s": 6.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5639025413684162, "source_start_sample": 7680, "source_end_sample": 337440, "first_word_output_sample": 3840, "last_word_output_sample": 322560, "samples": 329760, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 6.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u00-c00.mp3" }, "en-conversation-04-podcast-segment-timed-u03-c00": { "id": "en-conversation-04-podcast-segment-timed-u03-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76149, "plain": "Thank you. I was about to ask it to fit my entire life into three seconds.", "source_duration_s": 5.44, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 253440, "script_start_s": 0.24, "script_end_s": 5.28, "all_asr_last_word_s": 5.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4379697033588157, "source_start_sample": 7680, "source_end_sample": 260640, "first_word_output_sample": 3840, "last_word_output_sample": 245760, "samples": 252960, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-timed-u03-c00.mp3" }, "en-conversation-03-podcast-segment-free-u06-c00": { "id": "en-conversation-03-podcast-segment-free-u06-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75197, "plain": "References guide identity; acting directions guide expression. Neither is a magic guarantee.", "source_duration_s": 7.84, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 291840, "script_start_s": 0.16, "script_end_s": 6.08, "all_asr_last_word_s": 6.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.39444851236918166, "source_start_sample": 3840, "source_end_sample": 299040, "first_word_output_sample": 3840, "last_word_output_sample": 288000, "samples": 295200, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.61, "added_silence_s": 0.0, "duration_s": 6.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-free-u06-c00.mp3" }, "en-conversation-02-podcast-block-free-u01-c00": { "id": "en-conversation-02-podcast-block-free-u01-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 74617, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.28, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1109760, "script_start_s": 0.16, "script_end_s": 23.12, "all_asr_last_word_s": 23.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014285714285714285, "source_identity_reward": 0.46566610280045584, "source_start_sample": 3840, "source_end_sample": 1116960, "first_word_output_sample": 3840, "last_word_output_sample": 1105920, "samples": 1113120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u01-c00.mp3" }, "en-conversation-02-podcast-block-free-u01-c01": { "id": "en-conversation-02-podcast-block-free-u01-c01", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 74618, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.84, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1113600, "script_start_s": 0.0, "script_end_s": 23.2, "all_asr_last_word_s": 23.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014285714285714285, "source_identity_reward": 0.5369067399265747, "source_start_sample": 0, "source_end_sample": 1120800, "first_word_output_sample": 0, "last_word_output_sample": 1113600, "samples": 1120800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.49, "added_silence_s": 0.0, "duration_s": 23.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u01-c01.mp3" }, "en-conversation-02-podcast-block-free-u01-c02": { "id": "en-conversation-02-podcast-block-free-u01-c02", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 74619, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.76, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1105920, "script_start_s": 0.16, "script_end_s": 23.04, "all_asr_last_word_s": 23.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014285714285714285, "source_identity_reward": 0.463487343272811, "source_start_sample": 3840, "source_end_sample": 1113120, "first_word_output_sample": 3840, "last_word_output_sample": 1102080, "samples": 1109280, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.57, "added_silence_s": 0.0, "duration_s": 23.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u01-c02.mp3" }, "en-conversation-02-podcast-block-free-u01-c03": { "id": "en-conversation-02-podcast-block-free-u01-c03", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 74620, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 21.68, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1032960, "script_start_s": 0.16, "script_end_s": 21.52, "all_asr_last_word_s": 21.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014285714285714285, "source_identity_reward": 0.49967705960291126, "source_start_sample": 3840, "source_end_sample": 1040160, "first_word_output_sample": 3840, "last_word_output_sample": 1029120, "samples": 1036320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u01-c03.mp3" }, "en-conversation-02-podcast-block-free-u01-c04": { "id": "en-conversation-02-podcast-block-free-u01-c04", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 74621, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 32.08, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1113600, "script_start_s": 0.16, "script_end_s": 23.2, "all_asr_last_word_s": 31.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.11428571428571428, "source_identity_reward": 0.4105391435520805, "source_start_sample": 3840, "source_end_sample": 1120800, "first_word_output_sample": 3840, "last_word_output_sample": 1109760, "samples": 1116960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 8.73, "added_silence_s": 0.0, "duration_s": 23.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u01-c04.mp3" }, "en-conversation-02-podcast-block-free-u01-c05": { "id": "en-conversation-02-podcast-block-free-u01-c05", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 74622, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 22.96, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1094400, "script_start_s": 0.16, "script_end_s": 22.8, "all_asr_last_word_s": 22.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5037207702281996, "source_start_sample": 3840, "source_end_sample": 1101600, "first_word_output_sample": 3840, "last_word_output_sample": 1090560, "samples": 1097760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u01-c05.mp3" }, "en-conversation-02-podcast-block-free-u01-c06": { "id": "en-conversation-02-podcast-block-free-u01-c06", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 74623, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 22.8, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1086720, "script_start_s": 0.16, "script_end_s": 22.64, "all_asr_last_word_s": 22.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4538840312538323, "source_start_sample": 3840, "source_end_sample": 1093920, "first_word_output_sample": 3840, "last_word_output_sample": 1082880, "samples": 1090080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u01-c06.mp3" }, "en-conversation-02-podcast-block-free-u01-c07": { "id": "en-conversation-02-podcast-block-free-u01-c07", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 74624, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.44, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1117440, "script_start_s": 0.0, "script_end_s": 23.28, "all_asr_last_word_s": 23.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4589468103243593, "source_start_sample": 0, "source_end_sample": 1124640, "first_word_output_sample": 0, "last_word_output_sample": 1117440, "samples": 1124640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u01-c07.mp3" }, "de-conversation-02-podcast-segment-timed-u08-c00": { "id": "de-conversation-02-podcast-segment-timed-u08-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 4.8, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 226560, "script_start_s": 0.0, "script_end_s": 4.72, "all_asr_last_word_s": 4.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6588350085601627, "source_start_sample": 0, "source_end_sample": 230400, "first_word_output_sample": 0, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-timed-u08-c00.mp3" }, "de-conversation-05-podcast-segment-timed-u09-c00": { "id": "de-conversation-05-podcast-segment-timed-u09-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 238080, "script_start_s": 0.16, "script_end_s": 4.96, "all_asr_last_word_s": 4.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.39081109001354664, "source_start_sample": 3840, "source_end_sample": 241920, "first_word_output_sample": 3840, "last_word_output_sample": 234240, "samples": 241440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-timed-u09-c00.mp3" }, "en-conversation-04-podcast-segment-timed-u00-c00": { "id": "en-conversation-04-podcast-segment-timed-u00-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76101, "plain": "I can write a requested length next to an individual sentence in the script.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 222720, "script_start_s": 0.0, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.531822760822287, "source_start_sample": 0, "source_end_sample": 226560, "first_word_output_sample": 0, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-timed-u00-c00.mp3" }, "de-conversation-01-podcast-segment-free-u08-c00": { "id": "de-conversation-01-podcast-segment-free-u08-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 645120, "script_start_s": 0.24, "script_end_s": 13.44, "all_asr_last_word_s": 13.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.694944950702483, "source_start_sample": 7680, "source_end_sample": 652320, "first_word_output_sample": 3840, "last_word_output_sample": 637440, "samples": 644640, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 16.41, "added_silence_s": 0.0, "duration_s": 13.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-free-u08-c00.mp3" }, "de-conversation-01-podcast-segment-timed-u09-c00": { "id": "de-conversation-01-podcast-segment-timed-u09-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 238080, "script_start_s": 0.0, "script_end_s": 4.96, "all_asr_last_word_s": 4.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3463624374516172, "source_start_sample": 0, "source_end_sample": 241920, "first_word_output_sample": 0, "last_word_output_sample": 238080, "samples": 245280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-timed-u09-c00.mp3" }, "de-conversation-05-podcast-segment-free-u02-c00": { "id": "de-conversation-05-podcast-segment-free-u02-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87133, "plain": "Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 10, "matched_tokens": 10, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 272640, "script_start_s": 0.16, "script_end_s": 5.68, "all_asr_last_word_s": 26.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.4, "source_identity_reward": 0.28481122373593765, "source_start_sample": 3840, "source_end_sample": 279840, "first_word_output_sample": 3840, "last_word_output_sample": 268800, "samples": 276000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 24.17, "added_silence_s": 0.0, "duration_s": 5.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u02-c00.mp3" }, "de-conversation-05-podcast-block-timed-u00-c00": { "id": "de-conversation-05-podcast-block-timed-u00-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 87601, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.4, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 45, "coverage": 0.9782608695652174, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 779520, "script_start_s": 0.0, "script_end_s": 16.24, "all_asr_last_word_s": 16.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.021739130434782608, "source_identity_reward": 0.7878176171583627, "source_start_sample": 0, "source_end_sample": 786720, "first_word_output_sample": 0, "last_word_output_sample": 779520, "samples": 786720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u00-c00.mp3" }, "de-conversation-05-podcast-block-timed-u00-c01": { "id": "de-conversation-05-podcast-block-timed-u00-c01", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 87602, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.4, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 43, "coverage": 0.9347826086956522, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 783360, "script_start_s": 0.0, "script_end_s": 16.32, "all_asr_last_word_s": 16.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06521739130434782, "source_identity_reward": 0.7986607015723463, "source_start_sample": 0, "source_end_sample": 787200, "first_word_output_sample": 0, "last_word_output_sample": 783360, "samples": 790560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u00-c01.mp3" }, "de-conversation-05-podcast-block-timed-u00-c02": { "id": "de-conversation-05-podcast-block-timed-u00-c02", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 87603, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.4, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 45, "coverage": 0.9782608695652174, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 783360, "script_start_s": 0.0, "script_end_s": 16.32, "all_asr_last_word_s": 16.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.021739130434782608, "source_identity_reward": 0.8263092484674055, "source_start_sample": 0, "source_end_sample": 787200, "first_word_output_sample": 0, "last_word_output_sample": 783360, "samples": 790560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u00-c02.mp3" }, "de-conversation-05-podcast-block-timed-u00-c03": { "id": "de-conversation-05-podcast-block-timed-u00-c03", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 87604, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.32, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 42, "coverage": 0.9130434782608695, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 779520, "script_start_s": 0.0, "script_end_s": 16.24, "all_asr_last_word_s": 16.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08695652173913043, "source_identity_reward": 0.7585632873802097, "source_start_sample": 0, "source_end_sample": 783360, "first_word_output_sample": 0, "last_word_output_sample": 779520, "samples": 786720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u00-c03.mp3" }, "de-conversation-05-podcast-block-timed-u00-c04": { "id": "de-conversation-05-podcast-block-timed-u00-c04", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 87605, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.4, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 46, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 783360, "script_start_s": 0.0, "script_end_s": 16.32, "all_asr_last_word_s": 16.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7747990419537856, "source_start_sample": 0, "source_end_sample": 787200, "first_word_output_sample": 0, "last_word_output_sample": 783360, "samples": 790560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u00-c04.mp3" }, "de-conversation-05-podcast-block-timed-u00-c05": { "id": "de-conversation-05-podcast-block-timed-u00-c05", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 87606, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.4, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 43, "coverage": 0.9347826086956522, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 783360, "script_start_s": 0.0, "script_end_s": 16.32, "all_asr_last_word_s": 16.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06521739130434782, "source_identity_reward": 0.7636335872120326, "source_start_sample": 0, "source_end_sample": 787200, "first_word_output_sample": 0, "last_word_output_sample": 783360, "samples": 790560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u00-c05.mp3" }, "de-conversation-05-podcast-block-timed-u00-c06": { "id": "de-conversation-05-podcast-block-timed-u00-c06", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 87607, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.4, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 43, "coverage": 0.9347826086956522, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 783360, "script_start_s": 0.0, "script_end_s": 16.32, "all_asr_last_word_s": 16.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06521739130434782, "source_identity_reward": 0.7524797189806903, "source_start_sample": 0, "source_end_sample": 787200, "first_word_output_sample": 0, "last_word_output_sample": 783360, "samples": 790560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u00-c06.mp3" }, "de-conversation-05-podcast-block-timed-u00-c07": { "id": "de-conversation-05-podcast-block-timed-u00-c07", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 87608, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.4, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 44, "coverage": 0.9565217391304348, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.043478260869565216, "source_identity_reward": 0.770308426594923 }, "en-conversation-03-podcast-segment-free-u09-c00": { "id": "en-conversation-03-podcast-segment-free-u09-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 261120, "script_start_s": 0.16, "script_end_s": 5.44, "all_asr_last_word_s": 29.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.8333333333333333, "source_identity_reward": 0.08102002962774721, "source_start_sample": 3840, "source_end_sample": 268320, "first_word_output_sample": 3840, "last_word_output_sample": 257280, "samples": 264480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 24.41, "added_silence_s": 0.0, "duration_s": 5.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-free-u09-c00.mp3" }, "en-conversation-03-podcast-segment-timed-u04-c00": { "id": "en-conversation-03-podcast-segment-timed-u04-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75165, "plain": "We test that directly, turn by turn, rather than trusting the first sentence.", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 207360, "script_start_s": 0.24, "script_end_s": 4.32, "all_asr_last_word_s": 4.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3075100582244976, "source_start_sample": 7680, "source_end_sample": 211200, "first_word_output_sample": 3840, "last_word_output_sample": 199680, "samples": 206880, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-timed-u04-c00.mp3" }, "de-conversation-01-base17-segment-timed-u01-c00": { "id": "de-conversation-01-base17-segment-timed-u01-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83117, "plain": "Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht?", "source_duration_s": 3.6, "source_frame_cap": false, "alignment": { "expected_tokens": 10, "matched_tokens": 9, "coverage": 0.9, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 168960, "script_start_s": 0.08, "script_end_s": 3.52, "all_asr_last_word_s": 3.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1, "source_identity_reward": 0.557184737081169, "source_start_sample": 0, "source_end_sample": 172800, "first_word_output_sample": 3840, "last_word_output_sample": 168960, "samples": 176160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-timed-u01-c00.mp3" }, "de-conversation-04-base17-segment-timed-u07-c00": { "id": "de-conversation-04-base17-segment-timed-u07-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86213, "plain": "Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine.", "source_duration_s": 3.68, "source_frame_cap": false, "alignment": { "expected_tokens": 10, "matched_tokens": 8, "coverage": 0.8, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2, "source_identity_reward": 0.5479957840360138 }, "en-conversation-05-base17-segment-timed-u11-c00": { "id": "en-conversation-05-base17-segment-timed-u11-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 234240, "script_start_s": 0.16, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.0690701310312995, "source_start_sample": 3840, "source_end_sample": 238080, "first_word_output_sample": 3840, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-timed-u11-c00.mp3" }, "de-conversation-02-base17-segment-free-u02-c00": { "id": "de-conversation-02-base17-segment-free-u02-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84133, "plain": "Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 9, "coverage": 0.75, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 449280, "script_start_s": 0.16, "script_end_s": 9.36, "all_asr_last_word_s": 17.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.8333333333333334, "source_identity_reward": 0.32981715766332026, "source_start_sample": 3840, "source_end_sample": 456480, "first_word_output_sample": 3840, "last_word_output_sample": 445440, "samples": 452640, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 20.49, "added_silence_s": 0.0, "duration_s": 9.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-free-u02-c00.mp3" }, "en-conversation-02-base17-segment-free-u02-c00": { "id": "en-conversation-02-base17-segment-free-u02-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74133, "plain": "Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 12, "coverage": 0.9230769230769231, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 522240, "script_start_s": 0.0, "script_end_s": 10.88, "all_asr_last_word_s": 22.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.6153846153846154, "source_identity_reward": 0.1850401926533733, "source_start_sample": 0, "source_end_sample": 529440, "first_word_output_sample": 0, "last_word_output_sample": 522240, "samples": 529440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 18.97, "added_silence_s": 0.0, "duration_s": 11.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-free-u02-c00.mp3" }, "en-conversation-05-base17-segment-free-u02-c00": { "id": "en-conversation-05-base17-segment-free-u02-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77133, "plain": "That is what this five-language grid explores, with short and longer conversations.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 264960, "script_start_s": 0.0, "script_end_s": 5.52, "all_asr_last_word_s": 5.5200000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3850793965710617, "source_start_sample": 0, "source_end_sample": 272160, "first_word_output_sample": 0, "last_word_output_sample": 264960, "samples": 272160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.33, "added_silence_s": 0.0, "duration_s": 5.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-free-u02-c00.mp3" }, "de-conversation-05-base17-segment-free-u00-c00": { "id": "de-conversation-05-base17-segment-free-u00-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87101, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 345600, "script_start_s": 0.0, "script_end_s": 7.2, "all_asr_last_word_s": 28.080000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5454545454545454, "source_identity_reward": 0.37275273713359713, "source_start_sample": 0, "source_end_sample": 352800, "first_word_output_sample": 0, "last_word_output_sample": 345600, "samples": 352800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 22.65, "added_silence_s": 0.0, "duration_s": 7.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u00-c00.mp3" }, "en-conversation-02-base17-block-free-u00-c00": { "id": "en-conversation-02-base17-block-free-u00-c00", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 74601, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 67, "matched_tokens": 50, "coverage": 0.746268656716418, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2537313432835821, "source_identity_reward": 0.10799122250896645 }, "en-conversation-02-base17-block-free-u00-c01": { "id": "en-conversation-02-base17-block-free-u00-c01", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 74602, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1052160, "script_start_s": 0.16, "script_end_s": 21.92, "all_asr_last_word_s": 32.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04477611940298507, "source_identity_reward": 0.1665113713689047, "source_start_sample": 3840, "source_end_sample": 1059360, "first_word_output_sample": 3840, "last_word_output_sample": 1048320, "samples": 1055520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 37.93, "added_silence_s": 0.0, "duration_s": 21.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u00-c01.mp3" }, "en-conversation-02-base17-block-free-u00-c02": { "id": "en-conversation-02-base17-block-free-u00-c02", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 74603, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.64, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1075200, "script_start_s": 0.16, "script_end_s": 22.4, "all_asr_last_word_s": 22.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.12931915678737885, "source_start_sample": 3840, "source_end_sample": 1082400, "first_word_output_sample": 3840, "last_word_output_sample": 1071360, "samples": 1078560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 22.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u00-c02.mp3" }, "en-conversation-02-base17-block-free-u00-c03": { "id": "en-conversation-02-base17-block-free-u00-c03", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 74604, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 26.24, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1248000, "script_start_s": 0.24, "script_end_s": 26.0, "all_asr_last_word_s": 26.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.16158468001974124, "source_start_sample": 7680, "source_end_sample": 1255200, "first_word_output_sample": 3840, "last_word_output_sample": 1240320, "samples": 1247520, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 25.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u00-c03.mp3" }, "en-conversation-02-base17-block-free-u00-c04": { "id": "en-conversation-02-base17-block-free-u00-c04", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 74605, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 15360, "last_word_sample": 1176960, "script_start_s": 0.32, "script_end_s": 24.52, "all_asr_last_word_s": 24.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.14669620445175488, "source_start_sample": 11520, "source_end_sample": 1184160, "first_word_output_sample": 3840, "last_word_output_sample": 1165440, "samples": 1172640, "sample_rate": 48000, "head_removed_s": 0.24, "tail_removed_s": 35.33, "added_silence_s": 0.0, "duration_s": 24.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u00-c04.mp3" }, "en-conversation-02-base17-block-free-u00-c05": { "id": "en-conversation-02-base17-block-free-u00-c05", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 74606, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 67, "matched_tokens": 14, "coverage": 0.208955223880597, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.7910447761194029, "source_identity_reward": 0.01444853004793272 }, "en-conversation-02-base17-block-free-u00-c06": { "id": "en-conversation-02-base17-block-free-u00-c06", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 74607, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 52.8, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1059840, "script_start_s": 0.16, "script_end_s": 22.08, "all_asr_last_word_s": 48.879999999999995 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2835820895522388, "source_identity_reward": 0.15400382649261088, "source_start_sample": 3840, "source_end_sample": 1067040, "first_word_output_sample": 3840, "last_word_output_sample": 1056000, "samples": 1063200, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 30.57, "added_silence_s": 0.0, "duration_s": 22.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u00-c06.mp3" }, "en-conversation-02-base17-block-free-u00-c07": { "id": "en-conversation-02-base17-block-free-u00-c07", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 74608, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 67, "matched_tokens": 50, "coverage": 0.746268656716418, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2537313432835821, "source_identity_reward": 0.10100561874941186 }, "de-conversation-01-base17-block-free-u01-c00": { "id": "de-conversation-01-base17-block-free-u01-c00", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 83617, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 906240, "script_start_s": 0.16, "script_end_s": 18.88, "all_asr_last_word_s": 18.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.2968911587965761, "source_start_sample": 3840, "source_end_sample": 913440, "first_word_output_sample": 3840, "last_word_output_sample": 902400, "samples": 909600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.97, "added_silence_s": 0.0, "duration_s": 18.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u01-c00.mp3" }, "de-conversation-01-base17-block-free-u01-c01": { "id": "de-conversation-01-base17-block-free-u01-c01", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 83618, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 20.8, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 971520, "script_start_s": 0.24, "script_end_s": 20.24, "all_asr_last_word_s": 20.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04, "source_identity_reward": 0.42130500780162927, "source_start_sample": 7680, "source_end_sample": 978720, "first_word_output_sample": 3840, "last_word_output_sample": 963840, "samples": 971040, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.41, "added_silence_s": 0.0, "duration_s": 20.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u01-c01.mp3" }, "de-conversation-01-base17-block-free-u01-c02": { "id": "de-conversation-01-base17-block-free-u01-c02", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 83619, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1025280, "script_start_s": 0.24, "script_end_s": 21.36, "all_asr_last_word_s": 27.479999999999997 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3, "source_identity_reward": 0.21382634168631184, "source_start_sample": 7680, "source_end_sample": 1032480, "first_word_output_sample": 3840, "last_word_output_sample": 1017600, "samples": 1024800, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 38.49, "added_silence_s": 0.0, "duration_s": 21.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u01-c02.mp3" }, "de-conversation-01-base17-block-free-u01-c03": { "id": "de-conversation-01-base17-block-free-u01-c03", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 83620, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 975360, "script_start_s": 0.16, "script_end_s": 20.32, "all_asr_last_word_s": 40.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.36, "source_identity_reward": 0.2445265373364028, "source_start_sample": 3840, "source_end_sample": 982560, "first_word_output_sample": 3840, "last_word_output_sample": 971520, "samples": 978720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 39.53, "added_silence_s": 0.0, "duration_s": 20.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u01-c03.mp3" }, "de-conversation-01-base17-block-free-u01-c04": { "id": "de-conversation-01-base17-block-free-u01-c04", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 83621, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1029120, "script_start_s": 0.24, "script_end_s": 21.44, "all_asr_last_word_s": 21.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.21615889469006372, "source_start_sample": 7680, "source_end_sample": 1036320, "first_word_output_sample": 3840, "last_word_output_sample": 1021440, "samples": 1028640, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 38.41, "added_silence_s": 0.0, "duration_s": 21.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u01-c04.mp3" }, "de-conversation-01-base17-block-free-u01-c05": { "id": "de-conversation-01-base17-block-free-u01-c05", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 83622, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 32.48, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 50, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 15360, "last_word_sample": 1084800, "script_start_s": 0.32, "script_end_s": 22.6, "all_asr_last_word_s": 22.599999999999998 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2479812190656499, "source_start_sample": 11520, "source_end_sample": 1092000, "first_word_output_sample": 3840, "last_word_output_sample": 1073280, "samples": 1080480, "sample_rate": 48000, "head_removed_s": 0.24, "tail_removed_s": 9.73, "added_silence_s": 0.0, "duration_s": 22.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u01-c05.mp3" }, "de-conversation-01-base17-block-free-u01-c06": { "id": "de-conversation-01-base17-block-free-u01-c06", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 83623, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 50, "matched_tokens": 34, "coverage": 0.68, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.32, "source_identity_reward": 0.2172287031254134 }, "de-conversation-01-base17-block-free-u01-c07": { "id": "de-conversation-01-base17-block-free-u01-c07", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 83624, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 50, "matched_tokens": 48, "coverage": 0.96, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 998400, "script_start_s": 0.24, "script_end_s": 20.8, "all_asr_last_word_s": 20.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04, "source_identity_reward": 0.2531887432966506, "source_start_sample": 7680, "source_end_sample": 1005600, "first_word_output_sample": 3840, "last_word_output_sample": 990720, "samples": 997920, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 39.05, "added_silence_s": 0.0, "duration_s": 20.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u01-c07.mp3" }, "en-conversation-02-base17-segment-free-u06-c00": { "id": "en-conversation-02-base17-segment-free-u06-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74197, "plain": "That is why the reference clips sit beside the output. You can compare for yourself.", "source_duration_s": 11.68, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 245760, "script_start_s": 0.0, "script_end_s": 5.12, "all_asr_last_word_s": 5.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5357667250588158, "source_start_sample": 0, "source_end_sample": 252960, "first_word_output_sample": 0, "last_word_output_sample": 245760, "samples": 252960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 6.41, "added_silence_s": 0.0, "duration_s": 5.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-free-u06-c00.mp3" }, "de-conversation-01-base17-segment-free-u08-c00": { "id": "de-conversation-01-base17-segment-free-u08-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 23.92, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 11, "coverage": 0.8461538461538461, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 506880, "script_start_s": 0.16, "script_end_s": 10.56, "all_asr_last_word_s": 10.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.15384615384615385, "source_identity_reward": 0.6002138028988164, "source_start_sample": 3840, "source_end_sample": 514080, "first_word_output_sample": 3840, "last_word_output_sample": 503040, "samples": 510240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 13.21, "added_silence_s": 0.0, "duration_s": 10.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-free-u08-c00.mp3" }, "en-conversation-03-base17-segment-free-u09-c00": { "id": "en-conversation-03-base17-segment-free-u09-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 276480, "script_start_s": 0.16, "script_end_s": 5.76, "all_asr_last_word_s": 5.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2463523639315964, "source_start_sample": 3840, "source_end_sample": 283680, "first_word_output_sample": 3840, "last_word_output_sample": 272640, "samples": 279840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 24.09, "added_silence_s": 0.0, "duration_s": 5.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-free-u09-c00.mp3" }, "de-conversation-05-base17-segment-free-u08-c00": { "id": "de-conversation-05-base17-segment-free-u08-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 322560, "script_start_s": 0.0, "script_end_s": 6.72, "all_asr_last_word_s": 6.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6738930320186798, "source_start_sample": 0, "source_end_sample": 329760, "first_word_output_sample": 0, "last_word_output_sample": 322560, "samples": 329760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.13, "added_silence_s": 0.0, "duration_s": 6.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u08-c00.mp3" }, "en-conversation-04-base17-block-free-u02-c00": { "id": "en-conversation-04-base17-block-free-u02-c00", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 76633, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 12.56, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 526080, "script_start_s": 0.16, "script_end_s": 10.96, "all_asr_last_word_s": 10.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.10236421572758919, "source_start_sample": 3840, "source_end_sample": 533280, "first_word_output_sample": 3840, "last_word_output_sample": 522240, "samples": 529440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.45, "added_silence_s": 0.0, "duration_s": 11.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u02-c00.mp3" }, "en-conversation-04-base17-block-free-u02-c01": { "id": "en-conversation-04-base17-block-free-u02-c01", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 76634, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 487680, "script_start_s": 0.16, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2498062470157236, "source_start_sample": 3840, "source_end_sample": 494880, "first_word_output_sample": 3840, "last_word_output_sample": 483840, "samples": 491040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 49.69, "added_silence_s": 0.0, "duration_s": 10.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u02-c01.mp3" }, "en-conversation-04-base17-block-free-u02-c02": { "id": "en-conversation-04-base17-block-free-u02-c02", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 76635, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 11.92, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 480000, "script_start_s": 0.16, "script_end_s": 10.0, "all_asr_last_word_s": 10.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.11303583860269817, "source_start_sample": 3840, "source_end_sample": 487200, "first_word_output_sample": 3840, "last_word_output_sample": 476160, "samples": 483360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.77, "added_silence_s": 0.0, "duration_s": 10.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u02-c02.mp3" }, "en-conversation-04-base17-block-free-u02-c03": { "id": "en-conversation-04-base17-block-free-u02-c03", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 76636, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.28, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 437760, "script_start_s": 0.16, "script_end_s": 9.12, "all_asr_last_word_s": 9.120000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.20830465693759653, "source_start_sample": 3840, "source_end_sample": 444960, "first_word_output_sample": 3840, "last_word_output_sample": 433920, "samples": 441120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u02-c03.mp3" }, "en-conversation-04-base17-block-free-u02-c04": { "id": "en-conversation-04-base17-block-free-u02-c04", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 76637, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 468480, "script_start_s": 0.0, "script_end_s": 9.76, "all_asr_last_word_s": 16.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4827586206896552, "source_identity_reward": 0.10833401421773903, "source_start_sample": 0, "source_end_sample": 475680, "first_word_output_sample": 0, "last_word_output_sample": 468480, "samples": 475680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 50.09, "added_silence_s": 0.0, "duration_s": 9.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u02-c04.mp3" }, "en-conversation-04-base17-block-free-u02-c05": { "id": "en-conversation-04-base17-block-free-u02-c05", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 76638, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 453120, "script_start_s": 0.24, "script_end_s": 9.44, "all_asr_last_word_s": 53.040000000000006 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2413793103448276, "source_identity_reward": 0.0868276417263823, "source_start_sample": 7680, "source_end_sample": 460320, "first_word_output_sample": 3840, "last_word_output_sample": 445440, "samples": 452640, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 50.41, "added_silence_s": 0.0, "duration_s": 9.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u02-c05.mp3" }, "en-conversation-04-base17-block-free-u02-c06": { "id": "en-conversation-04-base17-block-free-u02-c06", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 76639, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 526080, "script_start_s": 0.0, "script_end_s": 10.96, "all_asr_last_word_s": 52.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0, "source_identity_reward": 0.13325597103101966, "source_start_sample": 0, "source_end_sample": 533280, "first_word_output_sample": 0, "last_word_output_sample": 526080, "samples": 533280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 48.89, "added_silence_s": 0.0, "duration_s": 11.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u02-c06.mp3" }, "en-conversation-04-base17-block-free-u02-c07": { "id": "en-conversation-04-base17-block-free-u02-c07", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 76640, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 12.72, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 472320, "script_start_s": 0.0, "script_end_s": 9.84, "all_asr_last_word_s": 9.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.22506452210397382, "source_start_sample": 0, "source_end_sample": 479520, "first_word_output_sample": 0, "last_word_output_sample": 472320, "samples": 479520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.73, "added_silence_s": 0.0, "duration_s": 9.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u02-c07.mp3" }, "en-conversation-04-base17-segment-timed-u10-c00": { "id": "en-conversation-04-base17-segment-timed-u10-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 218880, "script_start_s": 0.16, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.47931978001664743, "source_start_sample": 3840, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-timed-u10-c00.mp3" }, "en-conversation-03-base17-segment-timed-u05-c00": { "id": "en-conversation-03-base17-segment-timed-u05-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75181, "plain": "Good. I want the delighted host to remain the same person when she becomes doubtful.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 234240, "script_start_s": 0.24, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1901572716196698, "source_start_sample": 7680, "source_end_sample": 238080, "first_word_output_sample": 3840, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-timed-u05-c00.mp3" }, "en-conversation-05-base17-segment-free-u12-c00": { "id": "en-conversation-05-base17-segment-free-u12-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77293, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 314880, "script_start_s": 0.16, "script_end_s": 6.56, "all_asr_last_word_s": 24.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u12-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 2.6666666666666665, "source_identity_reward": 0.12090304330811162, "source_start_sample": 3840, "source_end_sample": 322080, "first_word_output_sample": 3840, "last_word_output_sample": 311040, "samples": 318240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.29, "added_silence_s": 0.0, "duration_s": 6.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-free-u12-c00.mp3" }, "de-conversation-04-base17-block-timed-u00-c00": { "id": "de-conversation-04-base17-block-timed-u00-c00", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 86601, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.84, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 61, "coverage": 0.9682539682539683, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1044480, "script_start_s": 0.0, "script_end_s": 21.76, "all_asr_last_word_s": 21.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.031746031746031744, "source_identity_reward": 0.23038027121112584, "source_start_sample": 0, "source_end_sample": 1048320, "first_word_output_sample": 0, "last_word_output_sample": 1044480, "samples": 1051680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u00-c00.mp3" }, "de-conversation-04-base17-block-timed-u00-c01": { "id": "de-conversation-04-base17-block-timed-u00-c01", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 86602, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 58, "coverage": 0.9206349206349206, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1040640, "script_start_s": 0.0, "script_end_s": 21.68, "all_asr_last_word_s": 21.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09523809523809523, "source_identity_reward": 0.2208135296790974, "source_start_sample": 0, "source_end_sample": 1044480, "first_word_output_sample": 0, "last_word_output_sample": 1040640, "samples": 1047840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u00-c01.mp3" }, "de-conversation-04-base17-block-timed-u00-c02": { "id": "de-conversation-04-base17-block-timed-u00-c02", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 86603, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 61, "coverage": 0.9682539682539683, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1040640, "script_start_s": 0.16, "script_end_s": 21.68, "all_asr_last_word_s": 21.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.031746031746031744, "source_identity_reward": 0.24227210807101182, "source_start_sample": 3840, "source_end_sample": 1044480, "first_word_output_sample": 3840, "last_word_output_sample": 1036800, "samples": 1044000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u00-c02.mp3" }, "de-conversation-04-base17-block-timed-u00-c03": { "id": "de-conversation-04-base17-block-timed-u00-c03", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 86604, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 61, "coverage": 0.9682539682539683, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1036800, "script_start_s": 0.16, "script_end_s": 21.6, "all_asr_last_word_s": 21.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.031746031746031744, "source_identity_reward": 0.20410673215522085, "source_start_sample": 3840, "source_end_sample": 1044000, "first_word_output_sample": 3840, "last_word_output_sample": 1032960, "samples": 1040160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u00-c03.mp3" }, "de-conversation-04-base17-block-timed-u00-c04": { "id": "de-conversation-04-base17-block-timed-u00-c04", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 86605, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 59, "coverage": 0.9365079365079365, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1040640, "script_start_s": 0.0, "script_end_s": 21.68, "all_asr_last_word_s": 21.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06349206349206349, "source_identity_reward": 0.3156522079027687, "source_start_sample": 0, "source_end_sample": 1044480, "first_word_output_sample": 0, "last_word_output_sample": 1040640, "samples": 1047840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u00-c04.mp3" }, "de-conversation-04-base17-block-timed-u00-c05": { "id": "de-conversation-04-base17-block-timed-u00-c05", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 86606, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 60, "coverage": 0.9523809523809523, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1040640, "script_start_s": 0.16, "script_end_s": 21.68, "all_asr_last_word_s": 21.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.047619047619047616, "source_identity_reward": 0.29211654379049257, "source_start_sample": 3840, "source_end_sample": 1044480, "first_word_output_sample": 3840, "last_word_output_sample": 1036800, "samples": 1044000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u00-c05.mp3" }, "de-conversation-04-base17-block-timed-u00-c06": { "id": "de-conversation-04-base17-block-timed-u00-c06", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 86607, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 61, "coverage": 0.9682539682539683, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1040640, "script_start_s": 0.0, "script_end_s": 21.68, "all_asr_last_word_s": 21.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.031746031746031744, "source_identity_reward": 0.24357355598911712, "source_start_sample": 0, "source_end_sample": 1044480, "first_word_output_sample": 0, "last_word_output_sample": 1040640, "samples": 1047840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u00-c06.mp3" }, "de-conversation-04-base17-block-timed-u00-c07": { "id": "de-conversation-04-base17-block-timed-u00-c07", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 86608, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 58, "coverage": 0.9206349206349206, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1036800, "script_start_s": 0.16, "script_end_s": 21.6, "all_asr_last_word_s": 21.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07936507936507936, "source_identity_reward": 0.32543991449922066, "source_start_sample": 3840, "source_end_sample": 1044000, "first_word_output_sample": 3840, "last_word_output_sample": 1032960, "samples": 1040160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u00-c07.mp3" }, "de-conversation-04-base17-segment-timed-u06-c00": { "id": "de-conversation-04-base17-segment-timed-u06-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86197, "plain": "Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 222720, "script_start_s": 0.16, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7424883934276125, "source_start_sample": 3840, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-timed-u06-c00.mp3" }, "en-conversation-05-base17-block-timed-u02-c00": { "id": "en-conversation-05-base17-block-timed-u02-c00", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 77633, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 53, "coverage": 0.9636363636363636, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03636363636363636, "source_identity_reward": 0.08036787255515183, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u02-c00.mp3" }, "en-conversation-05-base17-block-timed-u02-c01": { "id": "en-conversation-05-base17-block-timed-u02-c01", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 77634, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.11696586154989196, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u02-c01.mp3" }, "en-conversation-05-base17-block-timed-u02-c02": { "id": "en-conversation-05-base17-block-timed-u02-c02", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 77635, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.48, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 883200, "script_start_s": 0.16, "script_end_s": 18.4, "all_asr_last_word_s": 18.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1306284214022249, "source_start_sample": 3840, "source_end_sample": 887040, "first_word_output_sample": 3840, "last_word_output_sample": 879360, "samples": 886560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u02-c02.mp3" }, "en-conversation-05-base17-block-timed-u02-c03": { "id": "en-conversation-05-base17-block-timed-u02-c03", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 77636, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.09526223262409968, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u02-c03.mp3" }, "en-conversation-05-base17-block-timed-u02-c04": { "id": "en-conversation-05-base17-block-timed-u02-c04", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 77637, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 871680, "script_start_s": 0.16, "script_end_s": 18.16, "all_asr_last_word_s": 18.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.13588024439461222, "source_start_sample": 3840, "source_end_sample": 878880, "first_word_output_sample": 3840, "last_word_output_sample": 867840, "samples": 875040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 18.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u02-c04.mp3" }, "en-conversation-05-base17-block-timed-u02-c05": { "id": "en-conversation-05-base17-block-timed-u02-c05", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 77638, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.12503008380726546, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u02-c05.mp3" }, "en-conversation-05-base17-block-timed-u02-c06": { "id": "en-conversation-05-base17-block-timed-u02-c06", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 77639, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.11717884638111703, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u02-c06.mp3" }, "en-conversation-05-base17-block-timed-u02-c07": { "id": "en-conversation-05-base17-block-timed-u02-c07", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 77640, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 871680, "script_start_s": 0.16, "script_end_s": 18.16, "all_asr_last_word_s": 18.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.13224878477193494, "source_start_sample": 3840, "source_end_sample": 878880, "first_word_output_sample": 3840, "last_word_output_sample": 867840, "samples": 875040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 18.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u02-c07.mp3" }, "en-conversation-03-base17-block-free-u00-c00": { "id": "en-conversation-03-base17-block-free-u00-c00", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 75601, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 890880, "script_start_s": 0.0, "script_end_s": 18.56, "all_asr_last_word_s": 59.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.46153846153846156, "source_identity_reward": 0.11563515072784933, "source_start_sample": 0, "source_end_sample": 898080, "first_word_output_sample": 0, "last_word_output_sample": 890880, "samples": 898080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 41.29, "added_silence_s": 0.0, "duration_s": 18.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u00-c00.mp3" }, "en-conversation-03-base17-block-free-u00-c01": { "id": "en-conversation-03-base17-block-free-u00-c01", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 75602, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1029120, "script_start_s": 0.0, "script_end_s": 21.44, "all_asr_last_word_s": 59.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5769230769230769, "source_identity_reward": 0.10646458531972139, "source_start_sample": 0, "source_end_sample": 1036320, "first_word_output_sample": 0, "last_word_output_sample": 1029120, "samples": 1036320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.41, "added_silence_s": 0.0, "duration_s": 21.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u00-c01.mp3" }, "en-conversation-03-base17-block-free-u00-c02": { "id": "en-conversation-03-base17-block-free-u00-c02", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 75603, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 52, "matched_tokens": 25, "coverage": 0.4807692307692308, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5192307692307693, "source_identity_reward": 0.060306461182957155 }, "en-conversation-03-base17-block-free-u00-c03": { "id": "en-conversation-03-base17-block-free-u00-c03", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 75604, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 917760, "script_start_s": 0.16, "script_end_s": 19.12, "all_asr_last_word_s": 54.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.21153846153846154, "source_identity_reward": 0.22627020231623096, "source_start_sample": 3840, "source_end_sample": 924960, "first_word_output_sample": 3840, "last_word_output_sample": 913920, "samples": 921120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.73, "added_silence_s": 0.0, "duration_s": 19.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u00-c03.mp3" }, "en-conversation-03-base17-block-free-u00-c04": { "id": "en-conversation-03-base17-block-free-u00-c04", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 75605, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 919680, "script_start_s": 0.16, "script_end_s": 19.16, "all_asr_last_word_s": 19.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.1831152090737001, "source_start_sample": 3840, "source_end_sample": 926880, "first_word_output_sample": 3840, "last_word_output_sample": 915840, "samples": 923040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.69, "added_silence_s": 0.0, "duration_s": 19.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u00-c04.mp3" }, "en-conversation-03-base17-block-free-u00-c05": { "id": "en-conversation-03-base17-block-free-u00-c05", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 75606, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 21.6, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 948480, "script_start_s": 0.16, "script_end_s": 19.76, "all_asr_last_word_s": 19.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.19357982425905942, "source_start_sample": 3840, "source_end_sample": 955680, "first_word_output_sample": 3840, "last_word_output_sample": 944640, "samples": 951840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.69, "added_silence_s": 0.0, "duration_s": 19.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u00-c05.mp3" }, "en-conversation-03-base17-block-free-u00-c06": { "id": "en-conversation-03-base17-block-free-u00-c06", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 75607, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 52, "matched_tokens": 51, "coverage": 0.9807692307692307, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 954240, "script_start_s": 0.24, "script_end_s": 19.88, "all_asr_last_word_s": 19.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.019230769230769232, "source_identity_reward": 0.20812713536513294, "source_start_sample": 7680, "source_end_sample": 961440, "first_word_output_sample": 3840, "last_word_output_sample": 946560, "samples": 953760, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 39.97, "added_silence_s": 0.0, "duration_s": 19.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u00-c06.mp3" }, "en-conversation-03-base17-block-free-u00-c07": { "id": "en-conversation-03-base17-block-free-u00-c07", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 75608, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 52, "matched_tokens": 51, "coverage": 0.9807692307692307, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 946560, "script_start_s": 0.0, "script_end_s": 19.72, "all_asr_last_word_s": 42.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09615384615384616, "source_identity_reward": 0.18572821661456448, "source_start_sample": 0, "source_end_sample": 953760, "first_word_output_sample": 0, "last_word_output_sample": 946560, "samples": 953760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 40.13, "added_silence_s": 0.0, "duration_s": 19.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u00-c07.mp3" }, "de-conversation-05-base17-segment-timed-u15-c00": { "id": "de-conversation-05-base17-segment-timed-u15-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87341, "plain": "Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 6.8, "source_frame_cap": false, "alignment": { "expected_tokens": 19, "matched_tokens": 19, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 322560, "script_start_s": 0.16, "script_end_s": 6.72, "all_asr_last_word_s": 6.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u15-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.547141468543027, "source_start_sample": 3840, "source_end_sample": 326400, "first_word_output_sample": 3840, "last_word_output_sample": 318720, "samples": 325920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 6.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-timed-u15-c00.mp3" }, "en-conversation-04-base17-segment-timed-u04-c00": { "id": "en-conversation-04-base17-segment-timed-u04-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76165, "plain": "We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 192000, "script_start_s": 0.0, "script_end_s": 4.0, "all_asr_last_word_s": 4.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3952576364133194, "source_start_sample": 0, "source_end_sample": 195840, "first_word_output_sample": 0, "last_word_output_sample": 192000, "samples": 199200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-timed-u04-c00.mp3" }, "en-conversation-03-base17-segment-free-u01-c00": { "id": "en-conversation-03-base17-segment-free-u01-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75117, "plain": "Then keeping those voices distinct while the conversation wanders somewhere unexpected?", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 291840, "script_start_s": 0.16, "script_end_s": 6.08, "all_asr_last_word_s": 12.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.9090909090909091, "source_identity_reward": 0.13599631465431056, "source_start_sample": 3840, "source_end_sample": 299040, "first_word_output_sample": 3840, "last_word_output_sample": 288000, "samples": 295200, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.77, "added_silence_s": 0.0, "duration_s": 6.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-free-u01-c00.mp3" }, "en-conversation-03-base17-segment-free-u10-c00": { "id": "en-conversation-03-base17-segment-free-u10-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 245760, "script_start_s": 0.0, "script_end_s": 5.12, "all_asr_last_word_s": 5.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3757143194560911, "source_start_sample": 0, "source_end_sample": 252960, "first_word_output_sample": 0, "last_word_output_sample": 245760, "samples": 252960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.73, "added_silence_s": 0.0, "duration_s": 5.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-free-u10-c00.mp3" }, "en-conversation-03-base17-segment-timed-u02-c00": { "id": "en-conversation-03-base17-segment-timed-u02-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75133, "plain": "Yes. The model sees speaker-labelled references and, when available, earlier dialogue.", "source_duration_s": 3.76, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 11, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 176640, "script_start_s": 0.0, "script_end_s": 3.68, "all_asr_last_word_s": 3.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.38674726644498875, "source_start_sample": 0, "source_end_sample": 180480, "first_word_output_sample": 0, "last_word_output_sample": 176640, "samples": 183840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-timed-u02-c00.mp3" }, "de-conversation-01-base17-segment-free-u09-c00": { "id": "de-conversation-01-base17-segment-free-u09-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 299520, "script_start_s": 0.24, "script_end_s": 6.24, "all_asr_last_word_s": 6.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6054477031720475, "source_start_sample": 7680, "source_end_sample": 306720, "first_word_output_sample": 3840, "last_word_output_sample": 291840, "samples": 299040, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 23.61, "added_silence_s": 0.0, "duration_s": 6.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-free-u09-c00.mp3" }, "de-conversation-04-base17-block-timed-u02-c00": { "id": "de-conversation-04-base17-block-timed-u02-c00", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 86633, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 491520, "script_start_s": 0.0, "script_end_s": 10.24, "all_asr_last_word_s": 10.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2424355764943209, "source_start_sample": 0, "source_end_sample": 495360, "first_word_output_sample": 0, "last_word_output_sample": 491520, "samples": 498720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 10.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u02-c00.mp3" }, "de-conversation-04-base17-block-timed-u02-c01": { "id": "de-conversation-04-base17-block-timed-u02-c01", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 86634, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 491520, "script_start_s": 0.0, "script_end_s": 10.24, "all_asr_last_word_s": 10.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.22684373666735125, "source_start_sample": 0, "source_end_sample": 495360, "first_word_output_sample": 0, "last_word_output_sample": 491520, "samples": 498720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 10.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u02-c01.mp3" }, "de-conversation-04-base17-block-timed-u02-c02": { "id": "de-conversation-04-base17-block-timed-u02-c02", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 86635, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 491520, "script_start_s": 0.0, "script_end_s": 10.24, "all_asr_last_word_s": 10.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.47476591320283035, "source_start_sample": 0, "source_end_sample": 495360, "first_word_output_sample": 0, "last_word_output_sample": 491520, "samples": 498720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 10.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u02-c02.mp3" }, "de-conversation-04-base17-block-timed-u02-c03": { "id": "de-conversation-04-base17-block-timed-u02-c03", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 86636, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 491520, "script_start_s": 0.16, "script_end_s": 10.24, "all_asr_last_word_s": 10.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2221466470399619, "source_start_sample": 3840, "source_end_sample": 495360, "first_word_output_sample": 3840, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 10.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u02-c03.mp3" }, "de-conversation-04-base17-block-timed-u02-c04": { "id": "de-conversation-04-base17-block-timed-u02-c04", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 86637, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 487680, "script_start_s": 0.0, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.18020158773311618, "source_start_sample": 0, "source_end_sample": 494880, "first_word_output_sample": 0, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u02-c04.mp3" }, "de-conversation-04-base17-block-timed-u02-c05": { "id": "de-conversation-04-base17-block-timed-u02-c05", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 86638, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 491520, "script_start_s": 0.16, "script_end_s": 10.24, "all_asr_last_word_s": 10.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.39532725273306124, "source_start_sample": 3840, "source_end_sample": 495360, "first_word_output_sample": 3840, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 10.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u02-c05.mp3" }, "de-conversation-04-base17-block-timed-u02-c06": { "id": "de-conversation-04-base17-block-timed-u02-c06", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 86639, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 491520, "script_start_s": 0.16, "script_end_s": 10.24, "all_asr_last_word_s": 10.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.22147808471073924, "source_start_sample": 3840, "source_end_sample": 495360, "first_word_output_sample": 3840, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 10.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u02-c06.mp3" }, "de-conversation-04-base17-block-timed-u02-c07": { "id": "de-conversation-04-base17-block-timed-u02-c07", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 86640, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 26, "coverage": 0.896551724137931, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 491520, "script_start_s": 0.08, "script_end_s": 10.24, "all_asr_last_word_s": 10.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.10344827586206896, "source_identity_reward": 0.19703314041290068, "source_start_sample": 0, "source_end_sample": 495360, "first_word_output_sample": 3840, "last_word_output_sample": 491520, "samples": 498720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 10.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-timed-u02-c07.mp3" }, "en-conversation-05-base17-block-free-u03-c00": { "id": "en-conversation-05-base17-block-free-u03-c00", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 77649, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 23.52, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1063680, "script_start_s": 0.16, "script_end_s": 22.16, "all_asr_last_word_s": 22.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.09194751788238041, "source_start_sample": 3840, "source_end_sample": 1070880, "first_word_output_sample": 3840, "last_word_output_sample": 1059840, "samples": 1067040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.21, "added_silence_s": 0.0, "duration_s": 22.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u03-c00.mp3" }, "en-conversation-05-base17-block-free-u03-c01": { "id": "en-conversation-05-base17-block-free-u03-c01", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 77650, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1006080, "script_start_s": 0.0, "script_end_s": 20.96, "all_asr_last_word_s": 20.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u03-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.20282227276234194, "source_start_sample": 0, "source_end_sample": 1013280, "first_word_output_sample": 0, "last_word_output_sample": 1006080, "samples": 1013280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.89, "added_silence_s": 0.0, "duration_s": 21.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u03-c01.mp3" }, "en-conversation-05-base17-block-free-u03-c02": { "id": "en-conversation-05-base17-block-free-u03-c02", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 77651, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1000320, "script_start_s": 0.16, "script_end_s": 20.84, "all_asr_last_word_s": 36.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u03-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.11594202898550725, "source_identity_reward": 0.21184611240526696, "source_start_sample": 3840, "source_end_sample": 1007520, "first_word_output_sample": 3840, "last_word_output_sample": 996480, "samples": 1003680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 39.01, "added_silence_s": 0.0, "duration_s": 20.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u03-c02.mp3" }, "en-conversation-05-base17-block-free-u03-c03": { "id": "en-conversation-05-base17-block-free-u03-c03", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 77652, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1063680, "script_start_s": 0.24, "script_end_s": 22.16, "all_asr_last_word_s": 22.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u03-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.08891204671313903, "source_start_sample": 7680, "source_end_sample": 1070880, "first_word_output_sample": 3840, "last_word_output_sample": 1056000, "samples": 1063200, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 37.69, "added_silence_s": 0.0, "duration_s": 22.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u03-c03.mp3" }, "en-conversation-05-base17-block-free-u03-c04": { "id": "en-conversation-05-base17-block-free-u03-c04", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 77653, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 31.04, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1029120, "script_start_s": 0.16, "script_end_s": 21.44, "all_asr_last_word_s": 29.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u03-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2898550724637681, "source_identity_reward": 0.05942387613884614, "source_start_sample": 3840, "source_end_sample": 1036320, "first_word_output_sample": 3840, "last_word_output_sample": 1025280, "samples": 1032480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 9.45, "added_silence_s": 0.0, "duration_s": 21.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u03-c04.mp3" }, "en-conversation-05-base17-block-free-u03-c05": { "id": "en-conversation-05-base17-block-free-u03-c05", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 77654, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 37.92, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 936960, "script_start_s": 0.0, "script_end_s": 19.52, "all_asr_last_word_s": 26.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u03-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2028985507246377, "source_identity_reward": 0.11960187662298731, "source_start_sample": 0, "source_end_sample": 944160, "first_word_output_sample": 0, "last_word_output_sample": 936960, "samples": 944160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 18.25, "added_silence_s": 0.0, "duration_s": 19.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u03-c05.mp3" }, "en-conversation-05-base17-block-free-u03-c06": { "id": "en-conversation-05-base17-block-free-u03-c06", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 77655, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1121280, "script_start_s": 0.24, "script_end_s": 23.36, "all_asr_last_word_s": 23.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u03-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.1663401160187834, "source_start_sample": 7680, "source_end_sample": 1128480, "first_word_output_sample": 3840, "last_word_output_sample": 1113600, "samples": 1120800, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 36.49, "added_silence_s": 0.0, "duration_s": 23.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u03-c06.mp3" }, "en-conversation-05-base17-block-free-u03-c07": { "id": "en-conversation-05-base17-block-free-u03-c07", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 77656, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 22.8, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1029120, "script_start_s": 0.0, "script_end_s": 21.44, "all_asr_last_word_s": 21.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u03-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.12534288088834475, "source_start_sample": 0, "source_end_sample": 1036320, "first_word_output_sample": 0, "last_word_output_sample": 1029120, "samples": 1036320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.21, "added_silence_s": 0.0, "duration_s": 21.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u03-c07.mp3" }, "en-conversation-01-base17-segment-timed-u06-c00": { "id": "en-conversation-01-base17-segment-timed-u06-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73197, "plain": "The new training mixes real conversations with acting examples to retain both formats.", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 203520, "script_start_s": 0.0, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.42042553076371003, "source_start_sample": 0, "source_end_sample": 207360, "first_word_output_sample": 0, "last_word_output_sample": 203520, "samples": 210720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-timed-u06-c00.mp3" }, "en-conversation-05-base17-segment-timed-u02-c00": { "id": "en-conversation-05-base17-segment-timed-u02-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77133, "plain": "That is what this five-language grid explores, with short and longer conversations.", "source_duration_s": 4.0, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 184320, "script_start_s": 0.0, "script_end_s": 3.84, "all_asr_last_word_s": 3.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3450259757626497, "source_start_sample": 0, "source_end_sample": 191520, "first_word_output_sample": 0, "last_word_output_sample": 184320, "samples": 191520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 3.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-timed-u02-c00.mp3" }, "de-conversation-04-base17-segment-free-u08-c00": { "id": "de-conversation-04-base17-segment-free-u08-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 330240, "script_start_s": 0.08, "script_end_s": 6.88, "all_asr_last_word_s": 28.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.5384615384615385, "source_identity_reward": 0.25870614021881383, "source_start_sample": 0, "source_end_sample": 337440, "first_word_output_sample": 3840, "last_word_output_sample": 330240, "samples": 337440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 22.97, "added_silence_s": 0.0, "duration_s": 7.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-free-u08-c00.mp3" }, "en-conversation-04-base17-segment-free-u09-c00": { "id": "en-conversation-04-base17-segment-free-u09-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 291840, "script_start_s": 0.16, "script_end_s": 6.08, "all_asr_last_word_s": 6.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3483339521045401, "source_start_sample": 3840, "source_end_sample": 299040, "first_word_output_sample": 3840, "last_word_output_sample": 288000, "samples": 295200, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.77, "added_silence_s": 0.0, "duration_s": 6.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-free-u09-c00.mp3" }, "de-conversation-04-base17-segment-free-u03-c00": { "id": "de-conversation-04-base17-segment-free-u03-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86149, "plain": "Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 284160, "script_start_s": 0.24, "script_end_s": 5.92, "all_asr_last_word_s": 5.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6346228619658898, "source_start_sample": 7680, "source_end_sample": 291360, "first_word_output_sample": 3840, "last_word_output_sample": 276480, "samples": 283680, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 23.93, "added_silence_s": 0.0, "duration_s": 5.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-free-u03-c00.mp3" }, "en-conversation-01-base17-segment-free-u03-c00": { "id": "en-conversation-01-base17-segment-free-u03-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73149, "plain": "Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 257280, "script_start_s": 0.0, "script_end_s": 5.36, "all_asr_last_word_s": 12.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.9285714285714286, "source_identity_reward": 0.12758516398529543, "source_start_sample": 0, "source_end_sample": 264480, "first_word_output_sample": 0, "last_word_output_sample": 257280, "samples": 264480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.49, "added_silence_s": 0.0, "duration_s": 5.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-free-u03-c00.mp3" }, "de-conversation-02-base17-segment-free-u07-c00": { "id": "de-conversation-02-base17-segment-free-u07-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84213, "plain": "Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an.", "source_duration_s": 11.52, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 280320, "script_start_s": 0.16, "script_end_s": 5.84, "all_asr_last_word_s": 5.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6311332180652949, "source_start_sample": 3840, "source_end_sample": 287520, "first_word_output_sample": 3840, "last_word_output_sample": 276480, "samples": 283680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 5.53, "added_silence_s": 0.0, "duration_s": 5.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-free-u07-c00.mp3" }, "en-conversation-01-base17-segment-timed-u03-c00": { "id": "en-conversation-01-base17-segment-timed-u03-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73149, "plain": "Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 215040, "script_start_s": 0.16, "script_end_s": 4.48, "all_asr_last_word_s": 4.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.19596122806978553, "source_start_sample": 3840, "source_end_sample": 222240, "first_word_output_sample": 3840, "last_word_output_sample": 211200, "samples": 218400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-timed-u03-c00.mp3" }, "de-conversation-04-base17-segment-timed-u01-c00": { "id": "de-conversation-04-base17-segment-timed-u01-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86117, "plain": "Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde?", "source_duration_s": 3.92, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 184320, "script_start_s": 0.16, "script_end_s": 3.84, "all_asr_last_word_s": 3.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.547665201935518, "source_start_sample": 3840, "source_end_sample": 188160, "first_word_output_sample": 3840, "last_word_output_sample": 180480, "samples": 187680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-timed-u01-c00.mp3" }, "en-conversation-05-base17-segment-timed-u04-c00": { "id": "en-conversation-05-base17-segment-timed-u04-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77165, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers.", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 207360, "script_start_s": 0.0, "script_end_s": 4.32, "all_asr_last_word_s": 4.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4303820412618508, "source_start_sample": 0, "source_end_sample": 211200, "first_word_output_sample": 0, "last_word_output_sample": 207360, "samples": 214560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-timed-u04-c00.mp3" }, "de-conversation-04-base17-segment-timed-u03-c00": { "id": "de-conversation-04-base17-segment-timed-u03-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86149, "plain": "Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen.", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 203520, "script_start_s": 0.0, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.567066083720692, "source_start_sample": 0, "source_end_sample": 207360, "first_word_output_sample": 0, "last_word_output_sample": 203520, "samples": 210720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-timed-u03-c00.mp3" }, "de-conversation-05-base17-block-timed-u00-c00": { "id": "de-conversation-05-base17-block-timed-u00-c00", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 87601, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.4, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 45, "coverage": 0.9782608695652174, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 783360, "script_start_s": 0.16, "script_end_s": 16.32, "all_asr_last_word_s": 16.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.021739130434782608, "source_identity_reward": 0.2550819834470566, "source_start_sample": 3840, "source_end_sample": 787200, "first_word_output_sample": 3840, "last_word_output_sample": 779520, "samples": 786720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u00-c00.mp3" }, "de-conversation-05-base17-block-timed-u00-c01": { "id": "de-conversation-05-base17-block-timed-u00-c01", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 87602, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.48, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 45, "coverage": 0.9782608695652174, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 783360, "script_start_s": 0.0, "script_end_s": 16.32, "all_asr_last_word_s": 16.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.021739130434782608, "source_identity_reward": 0.2129686636098182, "source_start_sample": 0, "source_end_sample": 790560, "first_word_output_sample": 0, "last_word_output_sample": 783360, "samples": 790560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u00-c01.mp3" }, "de-conversation-05-base17-block-timed-u00-c02": { "id": "de-conversation-05-base17-block-timed-u00-c02", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 87603, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.56, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 45, "coverage": 0.9782608695652174, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 791040, "script_start_s": 0.0, "script_end_s": 16.48, "all_asr_last_word_s": 16.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.021739130434782608, "source_identity_reward": 0.22126297123906938, "source_start_sample": 0, "source_end_sample": 794880, "first_word_output_sample": 0, "last_word_output_sample": 791040, "samples": 798240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u00-c02.mp3" }, "de-conversation-05-base17-block-timed-u00-c03": { "id": "de-conversation-05-base17-block-timed-u00-c03", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 87604, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.48, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 45, "coverage": 0.9782608695652174, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 779520, "script_start_s": 0.0, "script_end_s": 16.24, "all_asr_last_word_s": 16.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.021739130434782608, "source_identity_reward": 0.2363601846534006, "source_start_sample": 0, "source_end_sample": 786720, "first_word_output_sample": 0, "last_word_output_sample": 779520, "samples": 786720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 16.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u00-c03.mp3" }, "de-conversation-05-base17-block-timed-u00-c04": { "id": "de-conversation-05-base17-block-timed-u00-c04", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 87605, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.48, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 43, "coverage": 0.9347826086956522, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 787200, "script_start_s": 0.16, "script_end_s": 16.4, "all_asr_last_word_s": 16.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06521739130434782, "source_identity_reward": 0.24411747080489005, "source_start_sample": 3840, "source_end_sample": 791040, "first_word_output_sample": 3840, "last_word_output_sample": 783360, "samples": 790560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u00-c04.mp3" }, "de-conversation-05-base17-block-timed-u00-c05": { "id": "de-conversation-05-base17-block-timed-u00-c05", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 87606, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.48, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 45, "coverage": 0.9782608695652174, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 787200, "script_start_s": 0.0, "script_end_s": 16.4, "all_asr_last_word_s": 16.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.021739130434782608, "source_identity_reward": 0.25098118622878024, "source_start_sample": 0, "source_end_sample": 791040, "first_word_output_sample": 0, "last_word_output_sample": 787200, "samples": 794400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u00-c05.mp3" }, "de-conversation-05-base17-block-timed-u00-c06": { "id": "de-conversation-05-base17-block-timed-u00-c06", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 87607, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.48, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 44, "coverage": 0.9565217391304348, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 787200, "script_start_s": 0.0, "script_end_s": 16.4, "all_asr_last_word_s": 16.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.043478260869565216, "source_identity_reward": 0.18606877906121527, "source_start_sample": 0, "source_end_sample": 791040, "first_word_output_sample": 0, "last_word_output_sample": 787200, "samples": 794400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u00-c06.mp3" }, "de-conversation-05-base17-block-timed-u00-c07": { "id": "de-conversation-05-base17-block-timed-u00-c07", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 87608, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch. Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus? Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen. Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 16.4, "source_frame_cap": false, "alignment": { "expected_tokens": 46, "matched_tokens": 44, "coverage": 0.9565217391304348, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 783360, "script_start_s": 0.0, "script_end_s": 16.32, "all_asr_last_word_s": 16.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.043478260869565216, "source_identity_reward": 0.30026958272312265, "source_start_sample": 0, "source_end_sample": 787200, "first_word_output_sample": 0, "last_word_output_sample": 783360, "samples": 790560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u00-c07.mp3" }, "en-conversation-02-base17-block-timed-u01-c00": { "id": "en-conversation-02-base17-block-timed-u01-c00", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 74617, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.28, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1113600, "script_start_s": 0.24, "script_end_s": 23.2, "all_asr_last_word_s": 23.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014285714285714285, "source_identity_reward": 0.1378150765854616, "source_start_sample": 7680, "source_end_sample": 1117440, "first_word_output_sample": 3840, "last_word_output_sample": 1105920, "samples": 1113120, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u01-c00.mp3" }, "en-conversation-02-base17-block-timed-u01-c01": { "id": "en-conversation-02-base17-block-timed-u01-c01", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 74618, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.36, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1113600, "script_start_s": 0.24, "script_end_s": 23.2, "all_asr_last_word_s": 23.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.09723723876437906, "source_start_sample": 7680, "source_end_sample": 1120800, "first_word_output_sample": 3840, "last_word_output_sample": 1105920, "samples": 1113120, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u01-c01.mp3" }, "en-conversation-02-base17-block-timed-u01-c02": { "id": "en-conversation-02-base17-block-timed-u01-c02", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 74619, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.36, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1117440, "script_start_s": 0.24, "script_end_s": 23.28, "all_asr_last_word_s": 23.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1860619772604391, "source_start_sample": 7680, "source_end_sample": 1121280, "first_word_output_sample": 3840, "last_word_output_sample": 1109760, "samples": 1116960, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u01-c02.mp3" }, "en-conversation-02-base17-block-timed-u01-c03": { "id": "en-conversation-02-base17-block-timed-u01-c03", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 74620, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.36, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1113600, "script_start_s": 0.16, "script_end_s": 23.2, "all_asr_last_word_s": 23.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014285714285714285, "source_identity_reward": 0.19287060538204748, "source_start_sample": 3840, "source_end_sample": 1120800, "first_word_output_sample": 3840, "last_word_output_sample": 1109760, "samples": 1116960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u01-c03.mp3" }, "en-conversation-02-base17-block-timed-u01-c04": { "id": "en-conversation-02-base17-block-timed-u01-c04", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 74621, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.36, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1117440, "script_start_s": 0.24, "script_end_s": 23.28, "all_asr_last_word_s": 23.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.15995993954054244, "source_start_sample": 7680, "source_end_sample": 1121280, "first_word_output_sample": 3840, "last_word_output_sample": 1109760, "samples": 1116960, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u01-c04.mp3" }, "en-conversation-02-base17-block-timed-u01-c05": { "id": "en-conversation-02-base17-block-timed-u01-c05", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 74622, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.36, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1117440, "script_start_s": 0.24, "script_end_s": 23.28, "all_asr_last_word_s": 23.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2628830433550815, "source_start_sample": 7680, "source_end_sample": 1121280, "first_word_output_sample": 3840, "last_word_output_sample": 1109760, "samples": 1116960, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u01-c05.mp3" }, "en-conversation-02-base17-block-timed-u01-c06": { "id": "en-conversation-02-base17-block-timed-u01-c06", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 74623, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.36, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1117440, "script_start_s": 0.24, "script_end_s": 23.28, "all_asr_last_word_s": 23.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2523261842577704, "source_start_sample": 7680, "source_end_sample": 1121280, "first_word_output_sample": 3840, "last_word_output_sample": 1109760, "samples": 1116960, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u01-c06.mp3" }, "en-conversation-02-base17-block-timed-u01-c07": { "id": "en-conversation-02-base17-block-timed-u01-c07", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 74624, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.36, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1117440, "script_start_s": 0.24, "script_end_s": 23.28, "all_asr_last_word_s": 23.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.23652413437158618, "source_start_sample": 7680, "source_end_sample": 1121280, "first_word_output_sample": 3840, "last_word_output_sample": 1109760, "samples": 1116960, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u01-c07.mp3" }, "de-conversation-04-base17-segment-timed-u11-c00": { "id": "de-conversation-04-base17-segment-timed-u11-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 218880, "script_start_s": 0.24, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6484133030055753, "source_start_sample": 7680, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 211200, "samples": 218400, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-timed-u11-c00.mp3" }, "de-conversation-04-base17-segment-free-u10-c00": { "id": "de-conversation-04-base17-segment-free-u10-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 299520, "script_start_s": 0.0, "script_end_s": 6.24, "all_asr_last_word_s": 6.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7519610164139641, "source_start_sample": 0, "source_end_sample": 306720, "first_word_output_sample": 0, "last_word_output_sample": 299520, "samples": 306720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 3.93, "added_silence_s": 0.0, "duration_s": 6.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-free-u10-c00.mp3" }, "en-conversation-03-base17-segment-timed-u10-c00": { "id": "en-conversation-03-base17-segment-timed-u10-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4315161091278905, "source_start_sample": 0, "source_end_sample": 222720, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-timed-u10-c00.mp3" }, "en-conversation-05-base17-segment-timed-u00-c00": { "id": "en-conversation-05-base17-segment-timed-u00-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77101, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch.", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 12, "coverage": 0.9230769230769231, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 203520, "script_start_s": 0.16, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07692307692307693, "source_identity_reward": 0.44531985378429606, "source_start_sample": 3840, "source_end_sample": 207360, "first_word_output_sample": 3840, "last_word_output_sample": 199680, "samples": 206880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-timed-u00-c00.mp3" }, "en-conversation-03-base17-segment-timed-u01-c00": { "id": "en-conversation-03-base17-segment-timed-u01-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75117, "plain": "Then keeping those voices distinct while the conversation wanders somewhere unexpected?", "source_duration_s": 3.76, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 176640, "script_start_s": 0.16, "script_end_s": 3.68, "all_asr_last_word_s": 3.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.18567373559750133, "source_start_sample": 3840, "source_end_sample": 180480, "first_word_output_sample": 3840, "last_word_output_sample": 172800, "samples": 180000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-timed-u01-c00.mp3" }, "en-conversation-02-base17-segment-timed-u04-c00": { "id": "en-conversation-02-base17-segment-timed-u04-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74165, "plain": "Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 218880, "script_start_s": 0.08, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.45851473621234506, "source_start_sample": 0, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-timed-u04-c00.mp3" }, "en-conversation-04-base17-segment-free-u02-c00": { "id": "en-conversation-04-base17-segment-free-u02-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76133, "plain": "No. The codec advances in eighty-millisecond frames, and requested timing can still miss.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 13, "coverage": 0.9285714285714286, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 268800, "script_start_s": 0.16, "script_end_s": 5.6, "all_asr_last_word_s": 26.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 2.5, "source_identity_reward": 0.13870945267138027, "source_start_sample": 3840, "source_end_sample": 276000, "first_word_output_sample": 3840, "last_word_output_sample": 264960, "samples": 272160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 24.25, "added_silence_s": 0.0, "duration_s": 5.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-free-u02-c00.mp3" }, "de-conversation-04-base17-segment-timed-u09-c00": { "id": "de-conversation-04-base17-segment-timed-u09-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 12, "coverage": 0.8571428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 234240, "script_start_s": 0.16, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.14285714285714285, "source_identity_reward": 0.4600862768064436, "source_start_sample": 3840, "source_end_sample": 238080, "first_word_output_sample": 3840, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-timed-u09-c00.mp3" }, "en-conversation-02-base17-segment-timed-u11-c00": { "id": "en-conversation-02-base17-segment-timed-u11-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 234240, "script_start_s": 0.24, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06666666666666667, "source_identity_reward": 0.19788733405814352, "source_start_sample": 7680, "source_end_sample": 238080, "first_word_output_sample": 3840, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-timed-u11-c00.mp3" }, "en-conversation-02-base17-segment-timed-u05-c00": { "id": "en-conversation-02-base17-segment-timed-u05-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74181, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 234240, "script_start_s": 0.24, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2452689016427818, "source_start_sample": 7680, "source_end_sample": 238080, "first_word_output_sample": 3840, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-timed-u05-c00.mp3" }, "en-conversation-01-base17-segment-free-u04-c00": { "id": "en-conversation-01-base17-segment-free-u04-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73165, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances.", "source_duration_s": 9.76, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 10, "coverage": 0.8333333333333334, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 253440, "script_start_s": 0.16, "script_end_s": 5.28, "all_asr_last_word_s": 5.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.16666666666666666, "source_identity_reward": 0.3853815066218141, "source_start_sample": 3840, "source_end_sample": 260640, "first_word_output_sample": 3840, "last_word_output_sample": 249600, "samples": 256800, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 4.33, "added_silence_s": 0.0, "duration_s": 5.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-free-u04-c00.mp3" }, "de-conversation-01-base17-block-timed-u00-c00": { "id": "de-conversation-01-base17-block-timed-u00-c00", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 83601, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.68, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 41, "coverage": 0.9318181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 748800, "script_start_s": 0.0, "script_end_s": 15.6, "all_asr_last_word_s": 15.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06818181818181818, "source_identity_reward": 0.25746842216846155, "source_start_sample": 0, "source_end_sample": 752640, "first_word_output_sample": 0, "last_word_output_sample": 748800, "samples": 756000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 15.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u00-c00.mp3" }, "de-conversation-01-base17-block-timed-u00-c01": { "id": "de-conversation-01-base17-block-timed-u00-c01", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 83602, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.68, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 41, "coverage": 0.9318181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 748800, "script_start_s": 0.08, "script_end_s": 15.6, "all_asr_last_word_s": 15.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06818181818181818, "source_identity_reward": 0.2757621986778095, "source_start_sample": 0, "source_end_sample": 752640, "first_word_output_sample": 3840, "last_word_output_sample": 748800, "samples": 756000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 15.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u00-c01.mp3" }, "de-conversation-01-base17-block-timed-u00-c02": { "id": "de-conversation-01-base17-block-timed-u00-c02", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 83603, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.68, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 42, "coverage": 0.9545454545454546, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 748800, "script_start_s": 0.0, "script_end_s": 15.6, "all_asr_last_word_s": 15.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.045454545454545456, "source_identity_reward": 0.18328743775833825, "source_start_sample": 0, "source_end_sample": 752640, "first_word_output_sample": 0, "last_word_output_sample": 748800, "samples": 756000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 15.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u00-c02.mp3" }, "de-conversation-01-base17-block-timed-u00-c03": { "id": "de-conversation-01-base17-block-timed-u00-c03", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 83604, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.68, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 42, "coverage": 0.9545454545454546, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 748800, "script_start_s": 0.0, "script_end_s": 15.6, "all_asr_last_word_s": 15.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.045454545454545456, "source_identity_reward": 0.26258511100370424, "source_start_sample": 0, "source_end_sample": 752640, "first_word_output_sample": 0, "last_word_output_sample": 748800, "samples": 756000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 15.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u00-c03.mp3" }, "de-conversation-01-base17-block-timed-u00-c04": { "id": "de-conversation-01-base17-block-timed-u00-c04", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 83605, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.68, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 42, "coverage": 0.9545454545454546, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 748800, "script_start_s": 0.0, "script_end_s": 15.6, "all_asr_last_word_s": 15.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.045454545454545456, "source_identity_reward": 0.27182480627790934, "source_start_sample": 0, "source_end_sample": 752640, "first_word_output_sample": 0, "last_word_output_sample": 748800, "samples": 756000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 15.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u00-c04.mp3" }, "de-conversation-01-base17-block-timed-u00-c05": { "id": "de-conversation-01-base17-block-timed-u00-c05", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 83606, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.68, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 39, "coverage": 0.8863636363636364, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 744960, "script_start_s": 0.0, "script_end_s": 15.52, "all_asr_last_word_s": 15.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.13636363636363635, "source_identity_reward": 0.1615189601633096, "source_start_sample": 0, "source_end_sample": 752160, "first_word_output_sample": 0, "last_word_output_sample": 744960, "samples": 752160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 15.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u00-c05.mp3" }, "de-conversation-01-base17-block-timed-u00-c06": { "id": "de-conversation-01-base17-block-timed-u00-c06", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 83607, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.68, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 42, "coverage": 0.9545454545454546, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 748800, "script_start_s": 0.0, "script_end_s": 15.6, "all_asr_last_word_s": 15.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.045454545454545456, "source_identity_reward": 0.2633093000357793, "source_start_sample": 0, "source_end_sample": 752640, "first_word_output_sample": 0, "last_word_output_sample": 748800, "samples": 756000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 15.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u00-c06.mp3" }, "de-conversation-01-base17-block-timed-u00-c07": { "id": "de-conversation-01-base17-block-timed-u00-c07", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 83608, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 15.68, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 40, "coverage": 0.9090909090909091, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 748800, "script_start_s": 0.0, "script_end_s": 15.6, "all_asr_last_word_s": 15.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.11363636363636363, "source_identity_reward": 0.29036472656071394, "source_start_sample": 0, "source_end_sample": 752640, "first_word_output_sample": 0, "last_word_output_sample": 748800, "samples": 756000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 15.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u00-c07.mp3" }, "en-conversation-03-base17-block-free-u01-c00": { "id": "en-conversation-03-base17-block-free-u01-c00", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 75617, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 17.76, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 837120, "script_start_s": 0.16, "script_end_s": 17.44, "all_asr_last_word_s": 17.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.14356035209348608, "source_start_sample": 3840, "source_end_sample": 844320, "first_word_output_sample": 3840, "last_word_output_sample": 833280, "samples": 840480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.17, "added_silence_s": 0.0, "duration_s": 17.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u01-c00.mp3" }, "en-conversation-03-base17-block-free-u01-c01": { "id": "en-conversation-03-base17-block-free-u01-c01", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 75618, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 848640, "script_start_s": 0.16, "script_end_s": 17.68, "all_asr_last_word_s": 17.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.10576379150213058, "source_start_sample": 3840, "source_end_sample": 855840, "first_word_output_sample": 3840, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 42.17, "added_silence_s": 0.0, "duration_s": 17.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u01-c01.mp3" }, "en-conversation-03-base17-block-free-u01-c02": { "id": "en-conversation-03-base17-block-free-u01-c02", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 75619, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 883200, "script_start_s": 0.0, "script_end_s": 18.4, "all_asr_last_word_s": 59.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5294117647058824, "source_identity_reward": 0.08894100778438284, "source_start_sample": 0, "source_end_sample": 890400, "first_word_output_sample": 0, "last_word_output_sample": 883200, "samples": 890400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 41.45, "added_silence_s": 0.0, "duration_s": 18.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u01-c02.mp3" }, "en-conversation-03-base17-block-free-u01-c03": { "id": "en-conversation-03-base17-block-free-u01-c03", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 75620, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 996480, "script_start_s": 0.24, "script_end_s": 20.76, "all_asr_last_word_s": 20.759999999999998 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.13018584322370602, "source_start_sample": 7680, "source_end_sample": 1003680, "first_word_output_sample": 3840, "last_word_output_sample": 988800, "samples": 996000, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 39.09, "added_silence_s": 0.0, "duration_s": 20.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u01-c03.mp3" }, "en-conversation-03-base17-block-free-u01-c04": { "id": "en-conversation-03-base17-block-free-u01-c04", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 75621, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 881280, "script_start_s": 0.0, "script_end_s": 18.36, "all_asr_last_word_s": 18.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.23066312524867744, "source_start_sample": 0, "source_end_sample": 888480, "first_word_output_sample": 0, "last_word_output_sample": 881280, "samples": 888480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 41.49, "added_silence_s": 0.0, "duration_s": 18.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u01-c04.mp3" }, "en-conversation-03-base17-block-free-u01-c05": { "id": "en-conversation-03-base17-block-free-u01-c05", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 75622, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 844800, "script_start_s": 0.0, "script_end_s": 17.6, "all_asr_last_word_s": 25.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0784313725490196, "source_identity_reward": 0.15052045309141063, "source_start_sample": 0, "source_end_sample": 852000, "first_word_output_sample": 0, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 42.25, "added_silence_s": 0.0, "duration_s": 17.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u01-c05.mp3" }, "en-conversation-03-base17-block-free-u01-c06": { "id": "en-conversation-03-base17-block-free-u01-c06", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 75623, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1021440, "script_start_s": 0.16, "script_end_s": 21.28, "all_asr_last_word_s": 21.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.08422904058341094, "source_start_sample": 3840, "source_end_sample": 1028640, "first_word_output_sample": 3840, "last_word_output_sample": 1017600, "samples": 1024800, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 38.57, "added_silence_s": 0.0, "duration_s": 21.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u01-c06.mp3" }, "en-conversation-03-base17-block-free-u01-c07": { "id": "en-conversation-03-base17-block-free-u01-c07", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 75624, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 998400, "script_start_s": 0.24, "script_end_s": 20.8, "all_asr_last_word_s": 20.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.14280495077086572, "source_start_sample": 7680, "source_end_sample": 1005600, "first_word_output_sample": 3840, "last_word_output_sample": 990720, "samples": 997920, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 39.05, "added_silence_s": 0.0, "duration_s": 20.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u01-c07.mp3" }, "de-conversation-01-base17-block-timed-u01-c00": { "id": "de-conversation-01-base17-block-timed-u01-c00", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 83617, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 18.0, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 48, "coverage": 0.96, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 860160, "script_start_s": 0.16, "script_end_s": 17.92, "all_asr_last_word_s": 17.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04, "source_identity_reward": 0.1999357719220165, "source_start_sample": 3840, "source_end_sample": 864000, "first_word_output_sample": 3840, "last_word_output_sample": 856320, "samples": 863520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u01-c00.mp3" }, "de-conversation-01-base17-block-timed-u01-c01": { "id": "de-conversation-01-base17-block-timed-u01-c01", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 83618, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 17.92, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 852480, "script_start_s": 0.24, "script_end_s": 17.76, "all_asr_last_word_s": 17.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.2494760568626417, "source_start_sample": 7680, "source_end_sample": 859680, "first_word_output_sample": 3840, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u01-c01.mp3" }, "de-conversation-01-base17-block-timed-u01-c02": { "id": "de-conversation-01-base17-block-timed-u01-c02", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 83619, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 17.92, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 852480, "script_start_s": 0.24, "script_end_s": 17.76, "all_asr_last_word_s": 17.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.2899778677778725, "source_start_sample": 7680, "source_end_sample": 859680, "first_word_output_sample": 3840, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u01-c02.mp3" }, "de-conversation-01-base17-block-timed-u01-c03": { "id": "de-conversation-01-base17-block-timed-u01-c03", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 83620, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 17.92, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 856320, "script_start_s": 0.24, "script_end_s": 17.84, "all_asr_last_word_s": 17.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.2795791618249832, "source_start_sample": 7680, "source_end_sample": 860160, "first_word_output_sample": 3840, "last_word_output_sample": 848640, "samples": 855840, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u01-c03.mp3" }, "de-conversation-01-base17-block-timed-u01-c04": { "id": "de-conversation-01-base17-block-timed-u01-c04", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 83621, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 17.92, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 856320, "script_start_s": 0.16, "script_end_s": 17.84, "all_asr_last_word_s": 17.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.2682594115913906, "source_start_sample": 3840, "source_end_sample": 860160, "first_word_output_sample": 3840, "last_word_output_sample": 852480, "samples": 859680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u01-c04.mp3" }, "de-conversation-01-base17-block-timed-u01-c05": { "id": "de-conversation-01-base17-block-timed-u01-c05", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 83622, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 17.92, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 848640, "script_start_s": 0.24, "script_end_s": 17.68, "all_asr_last_word_s": 17.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.2274065443762854, "source_start_sample": 7680, "source_end_sample": 855840, "first_word_output_sample": 3840, "last_word_output_sample": 840960, "samples": 848160, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 17.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u01-c05.mp3" }, "de-conversation-01-base17-block-timed-u01-c06": { "id": "de-conversation-01-base17-block-timed-u01-c06", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 83623, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 17.92, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 48, "coverage": 0.96, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 856320, "script_start_s": 0.24, "script_end_s": 17.84, "all_asr_last_word_s": 17.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04, "source_identity_reward": 0.2779317651108171, "source_start_sample": 7680, "source_end_sample": 860160, "first_word_output_sample": 3840, "last_word_output_sample": 848640, "samples": 855840, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u01-c06.mp3" }, "de-conversation-01-base17-block-timed-u01-c07": { "id": "de-conversation-01-base17-block-timed-u01-c07", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 83624, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 18.0, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 860160, "script_start_s": 0.16, "script_end_s": 17.92, "all_asr_last_word_s": 17.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.23274831053547357, "source_start_sample": 3840, "source_end_sample": 864000, "first_word_output_sample": 3840, "last_word_output_sample": 856320, "samples": 863520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-timed-u01-c07.mp3" }, "en-conversation-02-base17-segment-timed-u02-c00": { "id": "en-conversation-02-base17-segment-timed-u02-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74133, "plain": "Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 11, "coverage": 0.8461538461538461, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.15384615384615385, "source_identity_reward": 0.3911042298390389 }, "en-conversation-05-base17-segment-free-u04-c00": { "id": "en-conversation-05-base17-segment-free-u04-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77165, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 510720, "script_start_s": 0.0, "script_end_s": 10.64, "all_asr_last_word_s": 26.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.7692307692307692, "source_identity_reward": 0.18350350004277513, "source_start_sample": 0, "source_end_sample": 517920, "first_word_output_sample": 0, "last_word_output_sample": 510720, "samples": 517920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 19.21, "added_silence_s": 0.0, "duration_s": 10.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-free-u04-c00.mp3" }, "de-conversation-01-base17-segment-free-u02-c00": { "id": "de-conversation-01-base17-segment-free-u02-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83133, "plain": "Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 272640, "script_start_s": 0.0, "script_end_s": 5.68, "all_asr_last_word_s": 29.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 2.0, "source_identity_reward": 0.2179475863859098, "source_start_sample": 0, "source_end_sample": 279840, "first_word_output_sample": 0, "last_word_output_sample": 272640, "samples": 279840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.17, "added_silence_s": 0.0, "duration_s": 5.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-free-u02-c00.mp3" }, "de-conversation-05-base17-segment-timed-u12-c00": { "id": "de-conversation-05-base17-segment-timed-u12-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87293, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen.", "source_duration_s": 5.36, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 253440, "script_start_s": 0.0, "script_end_s": 5.28, "all_asr_last_word_s": 5.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u12-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6035287255999376, "source_start_sample": 0, "source_end_sample": 257280, "first_word_output_sample": 0, "last_word_output_sample": 253440, "samples": 260640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-timed-u12-c00.mp3" }, "de-conversation-05-base17-segment-timed-u06-c00": { "id": "de-conversation-05-base17-segment-timed-u06-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87197, "plain": "Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7459490611687749, "source_start_sample": 0, "source_end_sample": 226080, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-timed-u06-c00.mp3" }, "de-conversation-04-base17-block-free-u00-c00": { "id": "de-conversation-04-base17-block-free-u00-c00", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 86601, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 63, "matched_tokens": 60, "coverage": 0.9523809523809523, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1040640, "script_start_s": 0.08, "script_end_s": 21.68, "all_asr_last_word_s": 43.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.23809523809523808, "source_identity_reward": 0.1467898641703995, "source_start_sample": 0, "source_end_sample": 1047840, "first_word_output_sample": 3840, "last_word_output_sample": 1040640, "samples": 1047840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.17, "added_silence_s": 0.0, "duration_s": 21.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u00-c00.mp3" }, "de-conversation-04-base17-block-free-u00-c01": { "id": "de-conversation-04-base17-block-free-u00-c01", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 86602, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 63, "matched_tokens": 57, "coverage": 0.9047619047619048, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1219200, "script_start_s": 0.0, "script_end_s": 25.4, "all_asr_last_word_s": 25.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09523809523809523, "source_identity_reward": 0.3104704115235099, "source_start_sample": 0, "source_end_sample": 1226400, "first_word_output_sample": 0, "last_word_output_sample": 1219200, "samples": 1226400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 34.45, "added_silence_s": 0.0, "duration_s": 25.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u00-c01.mp3" }, "de-conversation-04-base17-block-free-u00-c02": { "id": "de-conversation-04-base17-block-free-u00-c02", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 86603, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 31.44, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 58, "coverage": 0.9206349206349206, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1073280, "script_start_s": 0.24, "script_end_s": 22.36, "all_asr_last_word_s": 22.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09523809523809523, "source_identity_reward": 0.3642031556802814, "source_start_sample": 7680, "source_end_sample": 1080480, "first_word_output_sample": 3840, "last_word_output_sample": 1065600, "samples": 1072800, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 8.93, "added_silence_s": 0.0, "duration_s": 22.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u00-c02.mp3" }, "de-conversation-04-base17-block-free-u00-c03": { "id": "de-conversation-04-base17-block-free-u00-c03", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 86604, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 63, "matched_tokens": 61, "coverage": 0.9682539682539683, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1009920, "script_start_s": 0.08, "script_end_s": 21.04, "all_asr_last_word_s": 21.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.031746031746031744, "source_identity_reward": 0.22753507499942127, "source_start_sample": 0, "source_end_sample": 1017120, "first_word_output_sample": 3840, "last_word_output_sample": 1009920, "samples": 1017120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.81, "added_silence_s": 0.0, "duration_s": 21.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u00-c03.mp3" }, "de-conversation-04-base17-block-free-u00-c04": { "id": "de-conversation-04-base17-block-free-u00-c04", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 86605, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 63, "matched_tokens": 44, "coverage": 0.6984126984126984, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.30158730158730157, "source_identity_reward": 0.1674804511490546 }, "de-conversation-04-base17-block-free-u00-c05": { "id": "de-conversation-04-base17-block-free-u00-c05", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 86606, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 63, "matched_tokens": 46, "coverage": 0.7301587301587301, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2698412698412698, "source_identity_reward": 0.2218109571824579 }, "de-conversation-04-base17-block-free-u00-c06": { "id": "de-conversation-04-base17-block-free-u00-c06", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 86607, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 22.64, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 59, "coverage": 0.9365079365079365, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1063680, "script_start_s": 0.0, "script_end_s": 22.16, "all_asr_last_word_s": 22.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06349206349206349, "source_identity_reward": 0.27819412938230376, "source_start_sample": 0, "source_end_sample": 1070880, "first_word_output_sample": 0, "last_word_output_sample": 1063680, "samples": 1070880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.33, "added_silence_s": 0.0, "duration_s": 22.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u00-c06.mp3" }, "de-conversation-04-base17-block-free-u00-c07": { "id": "de-conversation-04-base17-block-free-u00-c07", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 86608, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 22.72, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 58, "coverage": 0.9206349206349206, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1079040, "script_start_s": 0.16, "script_end_s": 22.48, "all_asr_last_word_s": 22.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07936507936507936, "source_identity_reward": 0.34457454272977023, "source_start_sample": 3840, "source_end_sample": 1086240, "first_word_output_sample": 3840, "last_word_output_sample": 1075200, "samples": 1082400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 22.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u00-c07.mp3" }, "en-conversation-03-base17-segment-timed-u09-c00": { "id": "en-conversation-03-base17-segment-timed-u09-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 4.0, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 188160, "script_start_s": 0.16, "script_end_s": 3.92, "all_asr_last_word_s": 3.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.07203987888631772, "source_start_sample": 3840, "source_end_sample": 192000, "first_word_output_sample": 3840, "last_word_output_sample": 184320, "samples": 191520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-timed-u09-c00.mp3" }, "de-conversation-04-base17-segment-free-u06-c00": { "id": "de-conversation-04-base17-segment-free-u06-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86197, "plain": "Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 345600, "script_start_s": 0.24, "script_end_s": 7.2, "all_asr_last_word_s": 7.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7238332083077137, "source_start_sample": 7680, "source_end_sample": 352800, "first_word_output_sample": 3840, "last_word_output_sample": 337920, "samples": 345120, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 22.65, "added_silence_s": 0.0, "duration_s": 7.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-free-u06-c00.mp3" }, "en-conversation-03-base17-block-free-u02-c00": { "id": "en-conversation-03-base17-block-free-u02-c00", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 75633, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 871680, "script_start_s": 0.0, "script_end_s": 18.16, "all_asr_last_word_s": 36.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2909090909090909, "source_identity_reward": 0.08415725416572276, "source_start_sample": 0, "source_end_sample": 878880, "first_word_output_sample": 0, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 41.69, "added_silence_s": 0.0, "duration_s": 18.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u02-c00.mp3" }, "en-conversation-03-base17-block-free-u02-c01": { "id": "en-conversation-03-base17-block-free-u02-c01", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 75634, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 860160, "script_start_s": 0.16, "script_end_s": 17.92, "all_asr_last_word_s": 35.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2545454545454545, "source_identity_reward": 0.07027921616865714, "source_start_sample": 3840, "source_end_sample": 867360, "first_word_output_sample": 3840, "last_word_output_sample": 856320, "samples": 863520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 41.93, "added_silence_s": 0.0, "duration_s": 17.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u02-c01.mp3" }, "en-conversation-03-base17-block-free-u02-c02": { "id": "en-conversation-03-base17-block-free-u02-c02", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 75635, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 848640, "script_start_s": 0.16, "script_end_s": 17.68, "all_asr_last_word_s": 17.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.12401520118078674, "source_start_sample": 3840, "source_end_sample": 855840, "first_word_output_sample": 3840, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 42.17, "added_silence_s": 0.0, "duration_s": 17.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u02-c02.mp3" }, "en-conversation-03-base17-block-free-u02-c03": { "id": "en-conversation-03-base17-block-free-u02-c03", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 75636, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 19.36, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.16639454516413732, "source_start_sample": 3840, "source_end_sample": 886560, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.89, "added_silence_s": 0.0, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u02-c03.mp3" }, "en-conversation-03-base17-block-free-u02-c04": { "id": "en-conversation-03-base17-block-free-u02-c04", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 75637, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 825600, "script_start_s": 0.16, "script_end_s": 17.2, "all_asr_last_word_s": 39.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.14545454545454545, "source_identity_reward": 0.1247235843886285, "source_start_sample": 3840, "source_end_sample": 832800, "first_word_output_sample": 3840, "last_word_output_sample": 821760, "samples": 828960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 42.65, "added_silence_s": 0.0, "duration_s": 17.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u02-c04.mp3" }, "en-conversation-03-base17-block-free-u02-c05": { "id": "en-conversation-03-base17-block-free-u02-c05", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 75638, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 52.48, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 875520, "script_start_s": 0.24, "script_end_s": 18.24, "all_asr_last_word_s": 52.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.38181818181818183, "source_identity_reward": 0.12165453016634943, "source_start_sample": 7680, "source_end_sample": 882720, "first_word_output_sample": 3840, "last_word_output_sample": 867840, "samples": 875040, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 34.09, "added_silence_s": 0.0, "duration_s": 18.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u02-c05.mp3" }, "en-conversation-03-base17-block-free-u02-c06": { "id": "en-conversation-03-base17-block-free-u02-c06", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 75639, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 877440, "script_start_s": 0.16, "script_end_s": 18.28, "all_asr_last_word_s": 18.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.10575680606280081, "source_start_sample": 3840, "source_end_sample": 884640, "first_word_output_sample": 3840, "last_word_output_sample": 873600, "samples": 880800, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 41.57, "added_silence_s": 0.0, "duration_s": 18.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u02-c06.mp3" }, "en-conversation-03-base17-block-free-u02-c07": { "id": "en-conversation-03-base17-block-free-u02-c07", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 75640, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 821760, "script_start_s": 0.0, "script_end_s": 17.12, "all_asr_last_word_s": 56.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5272727272727272, "source_identity_reward": 0.13229731771045217, "source_start_sample": 0, "source_end_sample": 828960, "first_word_output_sample": 0, "last_word_output_sample": 821760, "samples": 828960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 42.73, "added_silence_s": 0.0, "duration_s": 17.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-free-u02-c07.mp3" }, "en-conversation-04-base17-segment-timed-u06-c00": { "id": "en-conversation-04-base17-segment-timed-u06-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76197, "plain": "That is the idea: control the intention, measure the result, and listen carefully.", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 207360, "script_start_s": 0.16, "script_end_s": 4.32, "all_asr_last_word_s": 4.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5579735259570169, "source_start_sample": 3840, "source_end_sample": 211200, "first_word_output_sample": 3840, "last_word_output_sample": 203520, "samples": 210720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-timed-u06-c00.mp3" }, "en-conversation-05-base17-block-timed-u00-c00": { "id": "en-conversation-05-base17-block-timed-u00-c00", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 77601, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.08, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 864000, "script_start_s": 0.16, "script_end_s": 18.0, "all_asr_last_word_s": 18.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.13180845448422404, "source_start_sample": 3840, "source_end_sample": 867840, "first_word_output_sample": 3840, "last_word_output_sample": 860160, "samples": 867360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u00-c00.mp3" }, "en-conversation-05-base17-block-timed-u00-c01": { "id": "en-conversation-05-base17-block-timed-u00-c01", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 77602, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.0, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 860160, "script_start_s": 0.08, "script_end_s": 17.92, "all_asr_last_word_s": 17.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1873219146156163, "source_start_sample": 0, "source_end_sample": 864000, "first_word_output_sample": 3840, "last_word_output_sample": 860160, "samples": 867360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u00-c01.mp3" }, "en-conversation-05-base17-block-timed-u00-c02": { "id": "en-conversation-05-base17-block-timed-u00-c02", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 77603, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.08, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 864000, "script_start_s": 0.08, "script_end_s": 18.0, "all_asr_last_word_s": 18.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.12699239811574145, "source_start_sample": 0, "source_end_sample": 867840, "first_word_output_sample": 3840, "last_word_output_sample": 864000, "samples": 871200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u00-c02.mp3" }, "en-conversation-05-base17-block-timed-u00-c03": { "id": "en-conversation-05-base17-block-timed-u00-c03", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 77604, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.08, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 864000, "script_start_s": 0.08, "script_end_s": 18.0, "all_asr_last_word_s": 18.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.1246234745738935, "source_start_sample": 0, "source_end_sample": 867840, "first_word_output_sample": 3840, "last_word_output_sample": 864000, "samples": 871200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u00-c03.mp3" }, "en-conversation-05-base17-block-timed-u00-c04": { "id": "en-conversation-05-base17-block-timed-u00-c04", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 77605, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.08, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 864000, "script_start_s": 0.16, "script_end_s": 18.0, "all_asr_last_word_s": 18.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.12938257658275887, "source_start_sample": 3840, "source_end_sample": 867840, "first_word_output_sample": 3840, "last_word_output_sample": 860160, "samples": 867360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u00-c04.mp3" }, "en-conversation-05-base17-block-timed-u00-c05": { "id": "en-conversation-05-base17-block-timed-u00-c05", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 77606, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.0, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 860160, "script_start_s": 0.16, "script_end_s": 17.92, "all_asr_last_word_s": 17.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.17950611410615197, "source_start_sample": 3840, "source_end_sample": 864000, "first_word_output_sample": 3840, "last_word_output_sample": 856320, "samples": 863520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u00-c05.mp3" }, "en-conversation-05-base17-block-timed-u00-c06": { "id": "en-conversation-05-base17-block-timed-u00-c06", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 77607, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.08, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 864000, "script_start_s": 0.16, "script_end_s": 18.0, "all_asr_last_word_s": 18.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.25985089292034097, "source_start_sample": 3840, "source_end_sample": 867840, "first_word_output_sample": 3840, "last_word_output_sample": 860160, "samples": 867360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u00-c06.mp3" }, "en-conversation-05-base17-block-timed-u00-c07": { "id": "en-conversation-05-base17-block-timed-u00-c07", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 77608, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.0, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 860160, "script_start_s": 0.16, "script_end_s": 17.92, "all_asr_last_word_s": 17.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1916891496466069, "source_start_sample": 3840, "source_end_sample": 864000, "first_word_output_sample": 3840, "last_word_output_sample": 856320, "samples": 863520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u00-c07.mp3" }, "en-conversation-05-base17-segment-timed-u07-c00": { "id": "en-conversation-05-base17-segment-timed-u07-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77213, "plain": "All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 5.36, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 14, "coverage": 0.875, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.125, "source_identity_reward": 0.3048809254338515 }, "de-conversation-05-base17-segment-timed-u11-c00": { "id": "de-conversation-05-base17-segment-timed-u11-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 218880, "script_start_s": 0.24, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6465511954915877, "source_start_sample": 7680, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 211200, "samples": 218400, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-timed-u11-c00.mp3" }, "de-conversation-04-base17-block-free-u01-c00": { "id": "de-conversation-04-base17-block-free-u01-c00", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 86617, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 65, "matched_tokens": 48, "coverage": 0.7384615384615385, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.26153846153846155, "source_identity_reward": 0.2298946803750825 }, "de-conversation-04-base17-block-free-u01-c01": { "id": "de-conversation-04-base17-block-free-u01-c01", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 86618, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 31.28, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 61, "coverage": 0.9384615384615385, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1067520, "script_start_s": 0.16, "script_end_s": 22.24, "all_asr_last_word_s": 31.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.36923076923076925, "source_identity_reward": 0.2648661076596171, "source_start_sample": 3840, "source_end_sample": 1074720, "first_word_output_sample": 3840, "last_word_output_sample": 1063680, "samples": 1070880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 8.89, "added_silence_s": 0.0, "duration_s": 22.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u01-c01.mp3" }, "de-conversation-04-base17-block-free-u01-c02": { "id": "de-conversation-04-base17-block-free-u01-c02", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 86619, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 30.8, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 65, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1079040, "script_start_s": 0.16, "script_end_s": 22.48, "all_asr_last_word_s": 30.560000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2923076923076923, "source_identity_reward": 0.24784099547295826, "source_start_sample": 3840, "source_end_sample": 1086240, "first_word_output_sample": 3840, "last_word_output_sample": 1075200, "samples": 1082400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 8.17, "added_silence_s": 0.0, "duration_s": 22.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u01-c02.mp3" }, "de-conversation-04-base17-block-free-u01-c03": { "id": "de-conversation-04-base17-block-free-u01-c03", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 86620, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 65, "matched_tokens": 65, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1428480, "script_start_s": 0.0, "script_end_s": 29.76, "all_asr_last_word_s": 29.759999999999998 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.27692307692307694, "source_identity_reward": 0.2088428579305447, "source_start_sample": 0, "source_end_sample": 1435680, "first_word_output_sample": 0, "last_word_output_sample": 1428480, "samples": 1435680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 30.09, "added_silence_s": 0.0, "duration_s": 29.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u01-c03.mp3" }, "de-conversation-04-base17-block-free-u01-c04": { "id": "de-conversation-04-base17-block-free-u01-c04", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 86621, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 65, "matched_tokens": 65, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1207680, "script_start_s": 0.0, "script_end_s": 25.16, "all_asr_last_word_s": 25.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.015384615384615385, "source_identity_reward": 0.21148456067008217, "source_start_sample": 0, "source_end_sample": 1214880, "first_word_output_sample": 0, "last_word_output_sample": 1207680, "samples": 1214880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 34.69, "added_silence_s": 0.0, "duration_s": 25.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u01-c04.mp3" }, "de-conversation-04-base17-block-free-u01-c05": { "id": "de-conversation-04-base17-block-free-u01-c05", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 86622, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 65, "matched_tokens": 65, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1117440, "script_start_s": 0.16, "script_end_s": 23.28, "all_asr_last_word_s": 25.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.046153846153846156, "source_identity_reward": 0.2533951356696624, "source_start_sample": 3840, "source_end_sample": 1124640, "first_word_output_sample": 3840, "last_word_output_sample": 1113600, "samples": 1120800, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 36.57, "added_silence_s": 0.0, "duration_s": 23.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u01-c05.mp3" }, "de-conversation-04-base17-block-free-u01-c06": { "id": "de-conversation-04-base17-block-free-u01-c06", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 86623, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 65, "matched_tokens": 64, "coverage": 0.9846153846153847, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1209600, "script_start_s": 0.16, "script_end_s": 25.2, "all_asr_last_word_s": 43.120000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07692307692307693, "source_identity_reward": 0.25891805252221634, "source_start_sample": 3840, "source_end_sample": 1216800, "first_word_output_sample": 3840, "last_word_output_sample": 1205760, "samples": 1212960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 34.65, "added_silence_s": 0.0, "duration_s": 25.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u01-c06.mp3" }, "de-conversation-04-base17-block-free-u01-c07": { "id": "de-conversation-04-base17-block-free-u01-c07", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 86624, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 65, "matched_tokens": 64, "coverage": 0.9846153846153847, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1405440, "script_start_s": 0.16, "script_end_s": 29.28, "all_asr_last_word_s": 49.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.6, "source_identity_reward": 0.1313957884993934, "source_start_sample": 3840, "source_end_sample": 1412640, "first_word_output_sample": 3840, "last_word_output_sample": 1401600, "samples": 1408800, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 30.57, "added_silence_s": 0.0, "duration_s": 29.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u01-c07.mp3" }, "en-conversation-04-base17-block-free-u01-c00": { "id": "en-conversation-04-base17-block-free-u01-c00", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 76617, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 63, "matched_tokens": 62, "coverage": 0.9841269841269841, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 998400, "script_start_s": 0.08, "script_end_s": 20.8, "all_asr_last_word_s": 34.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.47619047619047616, "source_identity_reward": 0.12624666625488157, "source_start_sample": 0, "source_end_sample": 1005600, "first_word_output_sample": 3840, "last_word_output_sample": 998400, "samples": 1005600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 39.05, "added_silence_s": 0.0, "duration_s": 20.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u01-c00.mp3" }, "en-conversation-04-base17-block-free-u01-c01": { "id": "en-conversation-04-base17-block-free-u01-c01", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 76618, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 34.0, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 62, "coverage": 0.9841269841269841, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 948480, "script_start_s": 0.16, "script_end_s": 19.76, "all_asr_last_word_s": 28.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.31746031746031744, "source_identity_reward": 0.1447312829582618, "source_start_sample": 3840, "source_end_sample": 955680, "first_word_output_sample": 3840, "last_word_output_sample": 944640, "samples": 951840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 14.09, "added_silence_s": 0.0, "duration_s": 19.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u01-c01.mp3" }, "en-conversation-04-base17-block-free-u01-c02": { "id": "en-conversation-04-base17-block-free-u01-c02", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 76619, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1065600, "script_start_s": 0.16, "script_end_s": 22.2, "all_asr_last_word_s": 29.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.30158730158730157, "source_identity_reward": 0.1161017382828963, "source_start_sample": 3840, "source_end_sample": 1072800, "first_word_output_sample": 3840, "last_word_output_sample": 1061760, "samples": 1068960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 37.65, "added_silence_s": 0.0, "duration_s": 22.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u01-c02.mp3" }, "en-conversation-04-base17-block-free-u01-c03": { "id": "en-conversation-04-base17-block-free-u01-c03", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 76620, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 39.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 62, "coverage": 0.9841269841269841, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1034880, "script_start_s": 0.16, "script_end_s": 21.56, "all_asr_last_word_s": 29.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.31746031746031744, "source_identity_reward": 0.15134121891724842, "source_start_sample": 3840, "source_end_sample": 1042080, "first_word_output_sample": 3840, "last_word_output_sample": 1031040, "samples": 1038240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 18.05, "added_silence_s": 0.0, "duration_s": 21.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u01-c03.mp3" }, "en-conversation-04-base17-block-free-u01-c04": { "id": "en-conversation-04-base17-block-free-u01-c04", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 76621, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 936960, "script_start_s": 0.16, "script_end_s": 19.52, "all_asr_last_word_s": 51.959999999999994 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5714285714285714, "source_identity_reward": 0.10360860851212238, "source_start_sample": 3840, "source_end_sample": 944160, "first_word_output_sample": 3840, "last_word_output_sample": 933120, "samples": 940320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.33, "added_silence_s": 0.0, "duration_s": 19.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u01-c04.mp3" }, "en-conversation-04-base17-block-free-u01-c05": { "id": "en-conversation-04-base17-block-free-u01-c05", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 76622, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 35.28, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 933120, "script_start_s": 0.16, "script_end_s": 19.44, "all_asr_last_word_s": 24.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.19047619047619047, "source_identity_reward": 0.1272323956875435, "source_start_sample": 3840, "source_end_sample": 940320, "first_word_output_sample": 3840, "last_word_output_sample": 929280, "samples": 936480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 15.69, "added_silence_s": 0.0, "duration_s": 19.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u01-c05.mp3" }, "en-conversation-04-base17-block-free-u01-c06": { "id": "en-conversation-04-base17-block-free-u01-c06", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 76623, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 34.64, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 883200, "script_start_s": 0.0, "script_end_s": 18.4, "all_asr_last_word_s": 25.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.30158730158730157, "source_identity_reward": 0.19876399903034103, "source_start_sample": 0, "source_end_sample": 890400, "first_word_output_sample": 0, "last_word_output_sample": 883200, "samples": 890400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 16.09, "added_silence_s": 0.0, "duration_s": 18.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u01-c06.mp3" }, "en-conversation-04-base17-block-free-u01-c07": { "id": "en-conversation-04-base17-block-free-u01-c07", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 76624, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 63, "matched_tokens": 62, "coverage": 0.9841269841269841, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.19047619047619047, "source_identity_reward": 0.1503609063446375 }, "en-conversation-05-base17-segment-timed-u13-c00": { "id": "en-conversation-05-base17-segment-timed-u13-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77309, "plain": "And the second host is allowed to say, hold on, explain that again, I have not followed you.", "source_duration_s": 6.0, "source_frame_cap": false, "alignment": { "expected_tokens": 18, "matched_tokens": 18, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 280320, "script_start_s": 0.24, "script_end_s": 5.84, "all_asr_last_word_s": 5.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u13-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.29796785841867723, "source_start_sample": 7680, "source_end_sample": 287520, "first_word_output_sample": 3840, "last_word_output_sample": 272640, "samples": 279840, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-timed-u13-c00.mp3" }, "en-conversation-04-base17-segment-timed-u09-c00": { "id": "en-conversation-04-base17-segment-timed-u09-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 4.0, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 188160, "script_start_s": 0.16, "script_end_s": 3.92, "all_asr_last_word_s": 3.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.249849623989996, "source_start_sample": 3840, "source_end_sample": 192000, "first_word_output_sample": 3840, "last_word_output_sample": 184320, "samples": 191520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-timed-u09-c00.mp3" }, "de-conversation-01-base17-segment-free-u03-c00": { "id": "de-conversation-01-base17-segment-free-u03-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83149, "plain": "Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 253440, "script_start_s": 0.24, "script_end_s": 5.28, "all_asr_last_word_s": 21.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.2727272727272727, "source_identity_reward": 0.1713986290493314, "source_start_sample": 7680, "source_end_sample": 260640, "first_word_output_sample": 3840, "last_word_output_sample": 245760, "samples": 252960, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 24.57, "added_silence_s": 0.0, "duration_s": 5.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-free-u03-c00.mp3" }, "en-conversation-05-base17-segment-timed-u03-c00": { "id": "en-conversation-05-base17-segment-timed-u03-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77149, "plain": "I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 5.36, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 253440, "script_start_s": 0.16, "script_end_s": 5.28, "all_asr_last_word_s": 5.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.10443197888410667, "source_start_sample": 3840, "source_end_sample": 257280, "first_word_output_sample": 3840, "last_word_output_sample": 249600, "samples": 256800, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-timed-u03-c00.mp3" }, "en-conversation-03-base17-segment-timed-u04-c00": { "id": "en-conversation-03-base17-segment-timed-u04-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75165, "plain": "We test that directly, turn by turn, rather than trusting the first sentence.", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 203520, "script_start_s": 0.16, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.24628639631935612, "source_start_sample": 3840, "source_end_sample": 207360, "first_word_output_sample": 3840, "last_word_output_sample": 199680, "samples": 206880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-timed-u04-c00.mp3" }, "de-conversation-05-base17-segment-free-u09-c00": { "id": "de-conversation-05-base17-segment-free-u09-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 433920, "script_start_s": 0.16, "script_end_s": 9.04, "all_asr_last_word_s": 18.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.1428571428571428, "source_identity_reward": 0.2604613778993252, "source_start_sample": 3840, "source_end_sample": 441120, "first_word_output_sample": 3840, "last_word_output_sample": 430080, "samples": 437280, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 20.81, "added_silence_s": 0.0, "duration_s": 9.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u09-c00.mp3" }, "de-conversation-03-base17-block-timed-u02-c00": { "id": "de-conversation-03-base17-block-timed-u02-c00", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 85633, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 952320, "script_start_s": 0.24, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2574743266585486, "source_start_sample": 7680, "source_end_sample": 959520, "first_word_output_sample": 3840, "last_word_output_sample": 944640, "samples": 951840, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-timed-u02-c00.mp3" }, "de-conversation-03-base17-block-timed-u02-c01": { "id": "de-conversation-03-base17-block-timed-u02-c01", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 85634, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.08, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 54, "coverage": 0.9642857142857143, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 960000, "script_start_s": 0.08, "script_end_s": 20.0, "all_asr_last_word_s": 20.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.3425815498531425, "source_start_sample": 0, "source_end_sample": 963840, "first_word_output_sample": 3840, "last_word_output_sample": 960000, "samples": 967200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-timed-u02-c01.mp3" }, "de-conversation-03-base17-block-timed-u02-c02": { "id": "de-conversation-03-base17-block-timed-u02-c02", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 85635, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 952320, "script_start_s": 0.24, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.19593617357729426, "source_start_sample": 7680, "source_end_sample": 959520, "first_word_output_sample": 3840, "last_word_output_sample": 944640, "samples": 951840, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-timed-u02-c02.mp3" }, "de-conversation-03-base17-block-timed-u02-c03": { "id": "de-conversation-03-base17-block-timed-u02-c03", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 85636, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 54, "coverage": 0.9642857142857143, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 952320, "script_start_s": 0.16, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.15755104401537134, "source_start_sample": 3840, "source_end_sample": 959520, "first_word_output_sample": 3840, "last_word_output_sample": 948480, "samples": 955680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-timed-u02-c03.mp3" }, "de-conversation-03-base17-block-timed-u02-c04": { "id": "de-conversation-03-base17-block-timed-u02-c04", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 85637, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 956160, "script_start_s": 0.16, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1807124258645083, "source_start_sample": 3840, "source_end_sample": 960000, "first_word_output_sample": 3840, "last_word_output_sample": 952320, "samples": 959520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 19.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-timed-u02-c04.mp3" }, "de-conversation-03-base17-block-timed-u02-c05": { "id": "de-conversation-03-base17-block-timed-u02-c05", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 85638, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 952320, "script_start_s": 0.16, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.26889174422368256, "source_start_sample": 3840, "source_end_sample": 959520, "first_word_output_sample": 3840, "last_word_output_sample": 948480, "samples": 955680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-timed-u02-c05.mp3" }, "de-conversation-03-base17-block-timed-u02-c06": { "id": "de-conversation-03-base17-block-timed-u02-c06", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 85639, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3658911313210671, "source_start_sample": 0, "source_end_sample": 960000, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-timed-u02-c06.mp3" }, "de-conversation-03-base17-block-timed-u02-c07": { "id": "de-conversation-03-base17-block-timed-u02-c07", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 85640, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 54, "coverage": 0.9642857142857143, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 956160, "script_start_s": 0.16, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.2362737362730728, "source_start_sample": 3840, "source_end_sample": 960000, "first_word_output_sample": 3840, "last_word_output_sample": 952320, "samples": 959520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 19.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-timed-u02-c07.mp3" }, "en-conversation-05-base17-segment-timed-u12-c00": { "id": "en-conversation-05-base17-segment-timed-u12-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77293, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 230400, "script_start_s": 0.0, "script_end_s": 4.8, "all_asr_last_word_s": 4.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u12-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06666666666666667, "source_identity_reward": 0.4169143124802463, "source_start_sample": 0, "source_end_sample": 237600, "first_word_output_sample": 0, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-timed-u12-c00.mp3" }, "en-conversation-01-base17-segment-timed-u01-c00": { "id": "en-conversation-01-base17-segment-timed-u01-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73117, "plain": "You sound suspiciously excited. Did a speech model actually surprise you?", "source_duration_s": 3.76, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 172800, "script_start_s": 0.0, "script_end_s": 3.6, "all_asr_last_word_s": 3.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.13080935028627436, "source_start_sample": 0, "source_end_sample": 180000, "first_word_output_sample": 0, "last_word_output_sample": 172800, "samples": 180000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 3.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-timed-u01-c00.mp3" }, "de-conversation-03-podcast-segment-free-u10-c00": { "id": "de-conversation-03-podcast-segment-free-u10-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 19.6, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 314880, "script_start_s": 0.16, "script_end_s": 6.56, "all_asr_last_word_s": 19.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0, "source_identity_reward": 0.24910625628764035, "source_start_sample": 3840, "source_end_sample": 322080, "first_word_output_sample": 3840, "last_word_output_sample": 311040, "samples": 318240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 12.89, "added_silence_s": 0.0, "duration_s": 6.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-free-u10-c00.mp3" }, "de-conversation-02-podcast-segment-timed-u01-c00": { "id": "de-conversation-02-podcast-segment-timed-u01-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84117, "plain": "Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 192000, "script_start_s": 0.0, "script_end_s": 4.0, "all_asr_last_word_s": 4.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6900364430903805, "source_start_sample": 0, "source_end_sample": 195840, "first_word_output_sample": 0, "last_word_output_sample": 192000, "samples": 199200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-timed-u01-c00.mp3" }, "de-conversation-01-podcast-block-free-u02-c00": { "id": "de-conversation-01-podcast-block-free-u02-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 83633, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 53, "coverage": 0.9464285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1017600, "script_start_s": 0.0, "script_end_s": 21.2, "all_asr_last_word_s": 31.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.44642857142857145, "source_identity_reward": 0.5564732786565236, "source_start_sample": 0, "source_end_sample": 1024800, "first_word_output_sample": 0, "last_word_output_sample": 1017600, "samples": 1024800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.65, "added_silence_s": 0.0, "duration_s": 21.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u02-c00.mp3" }, "de-conversation-01-podcast-block-free-u02-c01": { "id": "de-conversation-01-podcast-block-free-u02-c01", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 83634, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 21.6, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1029120, "script_start_s": 0.0, "script_end_s": 21.44, "all_asr_last_word_s": 21.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6543148975509293, "source_start_sample": 0, "source_end_sample": 1036320, "first_word_output_sample": 0, "last_word_output_sample": 1029120, "samples": 1036320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u02-c01.mp3" }, "de-conversation-01-podcast-block-free-u02-c02": { "id": "de-conversation-01-podcast-block-free-u02-c02", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 83635, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.56, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 979200, "script_start_s": 0.0, "script_end_s": 20.4, "all_asr_last_word_s": 20.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.7763771197217608, "source_start_sample": 0, "source_end_sample": 986400, "first_word_output_sample": 0, "last_word_output_sample": 979200, "samples": 986400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u02-c02.mp3" }, "de-conversation-01-podcast-block-free-u02-c03": { "id": "de-conversation-01-podcast-block-free-u02-c03", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 83636, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 53, "coverage": 0.9464285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1006080, "script_start_s": 0.0, "script_end_s": 20.96, "all_asr_last_word_s": 59.480000000000004 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.44642857142857145, "source_identity_reward": 0.5965208858476516, "source_start_sample": 0, "source_end_sample": 1013280, "first_word_output_sample": 0, "last_word_output_sample": 1006080, "samples": 1013280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.89, "added_silence_s": 0.0, "duration_s": 21.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u02-c03.mp3" }, "de-conversation-01-podcast-block-free-u02-c04": { "id": "de-conversation-01-podcast-block-free-u02-c04", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 83637, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.64, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 983040, "script_start_s": 0.0, "script_end_s": 20.48, "all_asr_last_word_s": 20.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.017857142857142856, "source_identity_reward": 0.7762944654102244, "source_start_sample": 0, "source_end_sample": 990240, "first_word_output_sample": 0, "last_word_output_sample": 983040, "samples": 990240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u02-c04.mp3" }, "de-conversation-01-podcast-block-free-u02-c05": { "id": "de-conversation-01-podcast-block-free-u02-c05", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 83638, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.56, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 979200, "script_start_s": 0.0, "script_end_s": 20.4, "all_asr_last_word_s": 20.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.017857142857142856, "source_identity_reward": 0.7960961993525665, "source_start_sample": 0, "source_end_sample": 986400, "first_word_output_sample": 0, "last_word_output_sample": 979200, "samples": 986400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u02-c05.mp3" }, "de-conversation-01-podcast-block-free-u02-c06": { "id": "de-conversation-01-podcast-block-free-u02-c06", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 83639, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 27.04, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1006080, "script_start_s": 0.0, "script_end_s": 20.96, "all_asr_last_word_s": 26.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.23214285714285715, "source_identity_reward": 0.6891248815082707, "source_start_sample": 0, "source_end_sample": 1013280, "first_word_output_sample": 0, "last_word_output_sample": 1006080, "samples": 1013280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 5.93, "added_silence_s": 0.0, "duration_s": 21.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u02-c06.mp3" }, "de-conversation-01-podcast-block-free-u02-c07": { "id": "de-conversation-01-podcast-block-free-u02-c07", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 83640, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.48, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 975360, "script_start_s": 0.0, "script_end_s": 20.32, "all_asr_last_word_s": 20.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.017857142857142856, "source_identity_reward": 0.8223323152319679, "source_start_sample": 0, "source_end_sample": 982560, "first_word_output_sample": 0, "last_word_output_sample": 975360, "samples": 982560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u02-c07.mp3" }, "en-conversation-03-podcast-segment-timed-u05-c00": { "id": "en-conversation-03-podcast-segment-timed-u05-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75181, "plain": "Good. I want the delighted host to remain the same person when she becomes doubtful.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 238080, "script_start_s": 0.16, "script_end_s": 4.96, "all_asr_last_word_s": 4.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1126492354073248, "source_start_sample": 3840, "source_end_sample": 241920, "first_word_output_sample": 3840, "last_word_output_sample": 234240, "samples": 241440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-timed-u05-c00.mp3" }, "en-conversation-02-podcast-segment-free-u07-c00": { "id": "en-conversation-02-podcast-segment-free-u07-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74213, "plain": "Now that is a demonstration I can enjoy. Keep the little red light on.", "source_duration_s": 10.4, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 222720, "script_start_s": 0.0, "script_end_s": 4.64, "all_asr_last_word_s": 10.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2857142857142857, "source_identity_reward": 0.2617195156999984, "source_start_sample": 0, "source_end_sample": 229920, "first_word_output_sample": 0, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 5.61, "added_silence_s": 0.0, "duration_s": 4.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-free-u07-c00.mp3" }, "en-conversation-05-podcast-segment-free-u06-c00": { "id": "en-conversation-05-podcast-segment-free-u06-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77197, "plain": "Exactly. We publish the full scripts and seeds so the exciting failures remain visible too.", "source_duration_s": 10.64, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 264960, "script_start_s": 0.16, "script_end_s": 5.52, "all_asr_last_word_s": 5.5200000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06666666666666667, "source_identity_reward": 0.5393594154181771, "source_start_sample": 3840, "source_end_sample": 272160, "first_word_output_sample": 3840, "last_word_output_sample": 261120, "samples": 268320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 4.97, "added_silence_s": 0.0, "duration_s": 5.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-free-u06-c00.mp3" }, "en-conversation-02-podcast-segment-timed-u05-c00": { "id": "en-conversation-02-podcast-segment-timed-u05-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74181, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 238080, "script_start_s": 0.24, "script_end_s": 4.96, "all_asr_last_word_s": 4.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1880574286044237, "source_start_sample": 7680, "source_end_sample": 241920, "first_word_output_sample": 3840, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-timed-u05-c00.mp3" }, "de-conversation-04-podcast-segment-timed-u06-c00": { "id": "de-conversation-04-podcast-segment-timed-u06-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86197, "plain": "Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören.", "source_duration_s": 4.88, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 230400, "script_start_s": 0.0, "script_end_s": 4.8, "all_asr_last_word_s": 4.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5926873721498314, "source_start_sample": 0, "source_end_sample": 234240, "first_word_output_sample": 0, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-timed-u06-c00.mp3" }, "en-conversation-05-podcast-segment-free-u11-c00": { "id": "en-conversation-05-podcast-segment-free-u11-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.76, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 226560, "script_start_s": 0.08, "script_end_s": 4.72, "all_asr_last_word_s": 4.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.09170438231224833, "source_start_sample": 0, "source_end_sample": 233760, "first_word_output_sample": 3840, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.89, "added_silence_s": 0.0, "duration_s": 4.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-free-u11-c00.mp3" }, "en-conversation-05-podcast-block-timed-u01-c00": { "id": "en-conversation-05-podcast-block-timed-u01-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 77617, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.76, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 54, "coverage": 0.9152542372881356, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 940800, "script_start_s": 0.0, "script_end_s": 19.6, "all_asr_last_word_s": 19.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0847457627118644, "source_identity_reward": 0.4671891839638514, "source_start_sample": 0, "source_end_sample": 948000, "first_word_output_sample": 0, "last_word_output_sample": 940800, "samples": 948000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u01-c00.mp3" }, "en-conversation-05-podcast-block-timed-u01-c01": { "id": "en-conversation-05-podcast-block-timed-u01-c01", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 77618, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 936960, "script_start_s": 0.0, "script_end_s": 19.52, "all_asr_last_word_s": 19.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06779661016949153, "source_identity_reward": 0.44007800762588073, "source_start_sample": 0, "source_end_sample": 944160, "first_word_output_sample": 0, "last_word_output_sample": 936960, "samples": 944160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u01-c01.mp3" }, "en-conversation-05-podcast-block-timed-u01-c02": { "id": "en-conversation-05-podcast-block-timed-u01-c02", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 77619, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 936960, "script_start_s": 0.08, "script_end_s": 19.52, "all_asr_last_word_s": 19.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06779661016949153, "source_identity_reward": 0.5062501391385121, "source_start_sample": 0, "source_end_sample": 944160, "first_word_output_sample": 3840, "last_word_output_sample": 936960, "samples": 944160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u01-c02.mp3" }, "en-conversation-05-podcast-block-timed-u01-c03": { "id": "en-conversation-05-podcast-block-timed-u01-c03", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 77620, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 56, "coverage": 0.9491525423728814, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 940800, "script_start_s": 0.0, "script_end_s": 19.6, "all_asr_last_word_s": 19.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05084745762711865, "source_identity_reward": 0.4636151336144459, "source_start_sample": 0, "source_end_sample": 944640, "first_word_output_sample": 0, "last_word_output_sample": 940800, "samples": 948000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 19.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u01-c03.mp3" }, "en-conversation-05-podcast-block-timed-u01-c04": { "id": "en-conversation-05-podcast-block-timed-u01-c04", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 77621, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.76, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 54, "coverage": 0.9152542372881356, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 940800, "script_start_s": 0.0, "script_end_s": 19.6, "all_asr_last_word_s": 19.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0847457627118644, "source_identity_reward": 0.49421743117412487, "source_start_sample": 0, "source_end_sample": 948000, "first_word_output_sample": 0, "last_word_output_sample": 940800, "samples": 948000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u01-c04.mp3" }, "en-conversation-05-podcast-block-timed-u01-c05": { "id": "en-conversation-05-podcast-block-timed-u01-c05", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 77622, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 54, "coverage": 0.9152542372881356, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0847457627118644, "source_identity_reward": 0.48340155622473124 }, "en-conversation-05-podcast-block-timed-u01-c06": { "id": "en-conversation-05-podcast-block-timed-u01-c06", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 77623, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 54, "coverage": 0.9152542372881356, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 936960, "script_start_s": 0.0, "script_end_s": 19.52, "all_asr_last_word_s": 19.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0847457627118644, "source_identity_reward": 0.4372984252193294, "source_start_sample": 0, "source_end_sample": 944160, "first_word_output_sample": 0, "last_word_output_sample": 936960, "samples": 944160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u01-c06.mp3" }, "en-conversation-05-podcast-block-timed-u01-c07": { "id": "en-conversation-05-podcast-block-timed-u01-c07", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 77624, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 936960, "script_start_s": 0.0, "script_end_s": 19.52, "all_asr_last_word_s": 19.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06779661016949153, "source_identity_reward": 0.4744367600590462, "source_start_sample": 0, "source_end_sample": 944160, "first_word_output_sample": 0, "last_word_output_sample": 936960, "samples": 944160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u01-c07.mp3" }, "de-conversation-02-podcast-block-free-u00-c00": { "id": "de-conversation-02-podcast-block-free-u00-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 84601, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 24.32, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 57, "coverage": 0.95, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1117440, "script_start_s": 0.0, "script_end_s": 23.28, "all_asr_last_word_s": 23.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.7665985165067047, "source_start_sample": 0, "source_end_sample": 1124640, "first_word_output_sample": 0, "last_word_output_sample": 1117440, "samples": 1124640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.89, "added_silence_s": 0.0, "duration_s": 23.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u00-c00.mp3" }, "de-conversation-02-podcast-block-free-u00-c01": { "id": "de-conversation-02-podcast-block-free-u00-c01", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 84602, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 28.72, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 58, "coverage": 0.9666666666666667, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1125120, "script_start_s": 0.0, "script_end_s": 23.44, "all_asr_last_word_s": 28.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.21666666666666667, "source_identity_reward": 0.6584805282880928, "source_start_sample": 0, "source_end_sample": 1132320, "first_word_output_sample": 0, "last_word_output_sample": 1125120, "samples": 1132320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 5.13, "added_silence_s": 0.0, "duration_s": 23.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u00-c01.mp3" }, "de-conversation-02-podcast-block-free-u00-c02": { "id": "de-conversation-02-podcast-block-free-u00-c02", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 84603, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 25.2, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 54, "coverage": 0.9, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1148160, "script_start_s": 0.0, "script_end_s": 23.92, "all_asr_last_word_s": 23.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.11666666666666667, "source_identity_reward": 0.58852674558413, "source_start_sample": 0, "source_end_sample": 1155360, "first_word_output_sample": 0, "last_word_output_sample": 1148160, "samples": 1155360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.13, "added_silence_s": 0.0, "duration_s": 24.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u00-c02.mp3" }, "de-conversation-02-podcast-block-free-u00-c03": { "id": "de-conversation-02-podcast-block-free-u00-c03", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 84604, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 24.96, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 57, "coverage": 0.95, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1171200, "script_start_s": 0.0, "script_end_s": 24.4, "all_asr_last_word_s": 24.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06666666666666667, "source_identity_reward": 0.6278616492945817, "source_start_sample": 0, "source_end_sample": 1178400, "first_word_output_sample": 0, "last_word_output_sample": 1171200, "samples": 1178400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.41, "added_silence_s": 0.0, "duration_s": 24.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u00-c03.mp3" }, "de-conversation-02-podcast-block-free-u00-c04": { "id": "de-conversation-02-podcast-block-free-u00-c04", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 84605, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 25.36, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 57, "coverage": 0.95, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1171200, "script_start_s": 0.0, "script_end_s": 24.4, "all_asr_last_word_s": 24.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06666666666666667, "source_identity_reward": 0.6301947294009682, "source_start_sample": 0, "source_end_sample": 1178400, "first_word_output_sample": 0, "last_word_output_sample": 1171200, "samples": 1178400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.81, "added_silence_s": 0.0, "duration_s": 24.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u00-c04.mp3" }, "de-conversation-02-podcast-block-free-u00-c05": { "id": "de-conversation-02-podcast-block-free-u00-c05", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 84606, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 26.88, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 58, "coverage": 0.9666666666666667, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1121280, "script_start_s": 0.0, "script_end_s": 23.36, "all_asr_last_word_s": 23.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05, "source_identity_reward": 0.7863319788407709, "source_start_sample": 0, "source_end_sample": 1128480, "first_word_output_sample": 0, "last_word_output_sample": 1121280, "samples": 1128480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 3.37, "added_silence_s": 0.0, "duration_s": 23.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u00-c05.mp3" }, "de-conversation-02-podcast-block-free-u00-c06": { "id": "de-conversation-02-podcast-block-free-u00-c06", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 84607, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 24.4, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 58, "coverage": 0.9666666666666667, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1128960, "script_start_s": 0.0, "script_end_s": 23.52, "all_asr_last_word_s": 23.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05, "source_identity_reward": 0.7031604990465319, "source_start_sample": 0, "source_end_sample": 1136160, "first_word_output_sample": 0, "last_word_output_sample": 1128960, "samples": 1136160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.73, "added_silence_s": 0.0, "duration_s": 23.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u00-c06.mp3" }, "de-conversation-02-podcast-block-free-u00-c07": { "id": "de-conversation-02-podcast-block-free-u00-c07", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 84608, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 25.28, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 56, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1167360, "script_start_s": 0.0, "script_end_s": 24.32, "all_asr_last_word_s": 25.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1, "source_identity_reward": 0.7440979970433554, "source_start_sample": 0, "source_end_sample": 1174560, "first_word_output_sample": 0, "last_word_output_sample": 1167360, "samples": 1174560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.81, "added_silence_s": 0.0, "duration_s": 24.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u00-c07.mp3" }, "en-conversation-02-podcast-segment-timed-u07-c00": { "id": "en-conversation-02-podcast-segment-timed-u07-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74213, "plain": "Now that is a demonstration I can enjoy. Keep the little red light on.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 222720, "script_start_s": 0.16, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.06942943438467317, "source_start_sample": 3840, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-timed-u07-c00.mp3" }, "en-conversation-05-podcast-segment-timed-u03-c00": { "id": "en-conversation-05-podcast-segment-timed-u03-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77149, "plain": "I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 5.36, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 253440, "script_start_s": 0.0, "script_end_s": 5.28, "all_asr_last_word_s": 5.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.09904847860356362, "source_start_sample": 0, "source_end_sample": 257280, "first_word_output_sample": 0, "last_word_output_sample": 253440, "samples": 260640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-timed-u03-c00.mp3" }, "en-conversation-05-podcast-segment-free-u08-c00": { "id": "en-conversation-05-podcast-segment-free-u08-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 6.48, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 295680, "script_start_s": 0.16, "script_end_s": 6.16, "all_asr_last_word_s": 6.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.23588280888012764, "source_start_sample": 3840, "source_end_sample": 302880, "first_word_output_sample": 3840, "last_word_output_sample": 291840, "samples": 299040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.17, "added_silence_s": 0.0, "duration_s": 6.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-free-u08-c00.mp3" }, "en-conversation-04-podcast-segment-free-u10-c00": { "id": "en-conversation-04-podcast-segment-free-u10-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 11.44, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 264960, "script_start_s": 0.16, "script_end_s": 5.52, "all_asr_last_word_s": 10.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.7142857142857143, "source_identity_reward": 0.22967753370518684, "source_start_sample": 3840, "source_end_sample": 272160, "first_word_output_sample": 3840, "last_word_output_sample": 261120, "samples": 268320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 5.77, "added_silence_s": 0.0, "duration_s": 5.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-free-u10-c00.mp3" }, "de-conversation-02-podcast-segment-free-u09-c00": { "id": "de-conversation-02-podcast-segment-free-u09-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 314880, "script_start_s": 0.0, "script_end_s": 6.56, "all_asr_last_word_s": 14.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.9285714285714286, "source_identity_reward": 0.262255041938659, "source_start_sample": 0, "source_end_sample": 322080, "first_word_output_sample": 0, "last_word_output_sample": 314880, "samples": 322080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.29, "added_silence_s": 0.0, "duration_s": 6.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-free-u09-c00.mp3" }, "de-conversation-05-podcast-segment-free-u11-c00": { "id": "de-conversation-05-podcast-segment-free-u11-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 9.36, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 299520, "script_start_s": 0.16, "script_end_s": 6.24, "all_asr_last_word_s": 6.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.44812712876814254, "source_start_sample": 3840, "source_end_sample": 306720, "first_word_output_sample": 3840, "last_word_output_sample": 295680, "samples": 302880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 2.97, "added_silence_s": 0.0, "duration_s": 6.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u11-c00.mp3" }, "de-conversation-04-podcast-block-free-u01-c00": { "id": "de-conversation-04-podcast-block-free-u01-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 86617, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 38.16, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 65, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1201920, "script_start_s": 0.0, "script_end_s": 25.04, "all_asr_last_word_s": 38.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4153846153846154, "source_identity_reward": 0.5369998956469859, "source_start_sample": 0, "source_end_sample": 1209120, "first_word_output_sample": 0, "last_word_output_sample": 1201920, "samples": 1209120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 12.97, "added_silence_s": 0.0, "duration_s": 25.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u01-c00.mp3" }, "de-conversation-04-podcast-block-free-u01-c01": { "id": "de-conversation-04-podcast-block-free-u01-c01", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 86618, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 27.04, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 65, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1194240, "script_start_s": 0.0, "script_end_s": 24.88, "all_asr_last_word_s": 26.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06153846153846154, "source_identity_reward": 0.6775968896560186, "source_start_sample": 0, "source_end_sample": 1201440, "first_word_output_sample": 0, "last_word_output_sample": 1194240, "samples": 1201440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.01, "added_silence_s": 0.0, "duration_s": 25.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u01-c01.mp3" }, "de-conversation-04-podcast-block-free-u01-c02": { "id": "de-conversation-04-podcast-block-free-u01-c02", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 86619, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 24.72, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 65, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1178880, "script_start_s": 0.0, "script_end_s": 24.56, "all_asr_last_word_s": 24.560000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6957465252032178, "source_start_sample": 0, "source_end_sample": 1186080, "first_word_output_sample": 0, "last_word_output_sample": 1178880, "samples": 1186080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 24.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u01-c02.mp3" }, "de-conversation-04-podcast-block-free-u01-c03": { "id": "de-conversation-04-podcast-block-free-u01-c03", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 86620, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 46.8, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 65, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1175040, "script_start_s": 0.0, "script_end_s": 24.48, "all_asr_last_word_s": 46.68000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3384615384615385, "source_identity_reward": 0.5802418878568314, "source_start_sample": 0, "source_end_sample": 1182240, "first_word_output_sample": 0, "last_word_output_sample": 1175040, "samples": 1182240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 22.17, "added_silence_s": 0.0, "duration_s": 24.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u01-c03.mp3" }, "de-conversation-04-podcast-block-free-u01-c04": { "id": "de-conversation-04-podcast-block-free-u01-c04", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 86621, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 24.24, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 64, "coverage": 0.9846153846153847, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.015384615384615385, "source_identity_reward": 0.7422813138018602 }, "de-conversation-04-podcast-block-free-u01-c05": { "id": "de-conversation-04-podcast-block-free-u01-c05", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 86622, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 26.0, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 64, "coverage": 0.9846153846153847, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03076923076923077, "source_identity_reward": 0.8062340950012042 }, "de-conversation-04-podcast-block-free-u01-c06": { "id": "de-conversation-04-podcast-block-free-u01-c06", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 86623, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 24.48, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 65, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1167360, "script_start_s": 0.0, "script_end_s": 24.32, "all_asr_last_word_s": 24.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8097046647769799, "source_start_sample": 0, "source_end_sample": 1174560, "first_word_output_sample": 0, "last_word_output_sample": 1167360, "samples": 1174560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 24.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u01-c06.mp3" }, "de-conversation-04-podcast-block-free-u01-c07": { "id": "de-conversation-04-podcast-block-free-u01-c07", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 86624, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 32.48, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 63, "coverage": 0.9692307692307692, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1182720, "script_start_s": 0.0, "script_end_s": 24.64, "all_asr_last_word_s": 32.160000000000004 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3230769230769231, "source_identity_reward": 0.601832267483786, "source_start_sample": 0, "source_end_sample": 1189920, "first_word_output_sample": 0, "last_word_output_sample": 1182720, "samples": 1189920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 7.69, "added_silence_s": 0.0, "duration_s": 24.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u01-c07.mp3" }, "en-conversation-04-podcast-block-timed-u02-c00": { "id": "en-conversation-04-podcast-block-timed-u02-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 76633, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 456960, "script_start_s": 0.0, "script_end_s": 9.52, "all_asr_last_word_s": 9.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5333598584912512, "source_start_sample": 0, "source_end_sample": 464160, "first_word_output_sample": 0, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-timed-u02-c00.mp3" }, "en-conversation-04-podcast-block-timed-u02-c01": { "id": "en-conversation-04-podcast-block-timed-u02-c01", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 76634, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 456960, "script_start_s": 0.0, "script_end_s": 9.52, "all_asr_last_word_s": 9.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.45898875342471956, "source_start_sample": 0, "source_end_sample": 464160, "first_word_output_sample": 0, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-timed-u02-c01.mp3" }, "en-conversation-04-podcast-block-timed-u02-c02": { "id": "en-conversation-04-podcast-block-timed-u02-c02", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 76635, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 456960, "script_start_s": 0.0, "script_end_s": 9.52, "all_asr_last_word_s": 9.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5873817814476525, "source_start_sample": 0, "source_end_sample": 464160, "first_word_output_sample": 0, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-timed-u02-c02.mp3" }, "en-conversation-04-podcast-block-timed-u02-c03": { "id": "en-conversation-04-podcast-block-timed-u02-c03", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 76636, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 453120, "script_start_s": 0.16, "script_end_s": 9.44, "all_asr_last_word_s": 9.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3944411586908124, "source_start_sample": 3840, "source_end_sample": 460320, "first_word_output_sample": 3840, "last_word_output_sample": 449280, "samples": 456480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 9.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-timed-u02-c03.mp3" }, "en-conversation-04-podcast-block-timed-u02-c04": { "id": "en-conversation-04-podcast-block-timed-u02-c04", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 76637, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 453120, "script_start_s": 0.0, "script_end_s": 9.44, "all_asr_last_word_s": 9.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4396006670521705, "source_start_sample": 0, "source_end_sample": 460320, "first_word_output_sample": 0, "last_word_output_sample": 453120, "samples": 460320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 9.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-timed-u02-c04.mp3" }, "en-conversation-04-podcast-block-timed-u02-c05": { "id": "en-conversation-04-podcast-block-timed-u02-c05", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 76638, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 456960, "script_start_s": 0.0, "script_end_s": 9.52, "all_asr_last_word_s": 9.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3761924666789595, "source_start_sample": 0, "source_end_sample": 464160, "first_word_output_sample": 0, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-timed-u02-c05.mp3" }, "en-conversation-04-podcast-block-timed-u02-c06": { "id": "en-conversation-04-podcast-block-timed-u02-c06", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 76639, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 456960, "script_start_s": 0.0, "script_end_s": 9.52, "all_asr_last_word_s": 9.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4594260417490288, "source_start_sample": 0, "source_end_sample": 464160, "first_word_output_sample": 0, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-timed-u02-c06.mp3" }, "en-conversation-04-podcast-block-timed-u02-c07": { "id": "en-conversation-04-podcast-block-timed-u02-c07", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 76640, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 456960, "script_start_s": 0.0, "script_end_s": 9.52, "all_asr_last_word_s": 9.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4643650617351178, "source_start_sample": 0, "source_end_sample": 464160, "first_word_output_sample": 0, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-timed-u02-c07.mp3" }, "de-conversation-03-podcast-segment-timed-u10-c00": { "id": "de-conversation-03-podcast-segment-timed-u10-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 5.76, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 272640, "script_start_s": 0.08, "script_end_s": 5.68, "all_asr_last_word_s": 5.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6158368226573863, "source_start_sample": 0, "source_end_sample": 276480, "first_word_output_sample": 3840, "last_word_output_sample": 272640, "samples": 279840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-timed-u10-c00.mp3" }, "en-conversation-05-podcast-block-free-u03-c00": { "id": "en-conversation-05-podcast-block-free-u03-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 77649, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 21.84, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1040640, "script_start_s": 0.24, "script_end_s": 21.68, "all_asr_last_word_s": 21.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.5504569548123095, "source_start_sample": 7680, "source_end_sample": 1047840, "first_word_output_sample": 3840, "last_word_output_sample": 1032960, "samples": 1040160, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u03-c00.mp3" }, "en-conversation-05-podcast-block-free-u03-c01": { "id": "en-conversation-05-podcast-block-free-u03-c01", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 77650, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1044480, "script_start_s": 0.0, "script_end_s": 21.76, "all_asr_last_word_s": 59.480000000000004 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u03-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0434782608695652, "source_identity_reward": 0.2878650089497262, "source_start_sample": 0, "source_end_sample": 1051680, "first_word_output_sample": 0, "last_word_output_sample": 1044480, "samples": 1051680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.09, "added_silence_s": 0.0, "duration_s": 21.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u03-c01.mp3" }, "en-conversation-05-podcast-block-free-u03-c02": { "id": "en-conversation-05-podcast-block-free-u03-c02", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 77651, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 21.44, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1021440, "script_start_s": 0.16, "script_end_s": 21.28, "all_asr_last_word_s": 21.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u03-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.4717708491853493, "source_start_sample": 3840, "source_end_sample": 1028640, "first_word_output_sample": 3840, "last_word_output_sample": 1017600, "samples": 1024800, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u03-c02.mp3" }, "en-conversation-05-podcast-block-free-u03-c03": { "id": "en-conversation-05-podcast-block-free-u03-c03", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 77652, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 21.52, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1025280, "script_start_s": 0.0, "script_end_s": 21.36, "all_asr_last_word_s": 21.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u03-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.5411525880673871, "source_start_sample": 0, "source_end_sample": 1032480, "first_word_output_sample": 0, "last_word_output_sample": 1025280, "samples": 1032480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u03-c03.mp3" }, "en-conversation-05-podcast-block-free-u03-c04": { "id": "en-conversation-05-podcast-block-free-u03-c04", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 77653, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 25.36, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 998400, "script_start_s": 0.16, "script_end_s": 20.8, "all_asr_last_word_s": 24.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u03-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.13043478260869565, "source_identity_reward": 0.5188718286638156, "source_start_sample": 3840, "source_end_sample": 1005600, "first_word_output_sample": 3840, "last_word_output_sample": 994560, "samples": 1001760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 4.41, "added_silence_s": 0.0, "duration_s": 20.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u03-c04.mp3" }, "en-conversation-05-podcast-block-free-u03-c05": { "id": "en-conversation-05-podcast-block-free-u03-c05", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 77654, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 21.28, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 67, "coverage": 0.9710144927536232, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1013760, "script_start_s": 0.16, "script_end_s": 21.12, "all_asr_last_word_s": 21.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u03-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.028985507246376812, "source_identity_reward": 0.5935450785373935, "source_start_sample": 3840, "source_end_sample": 1020960, "first_word_output_sample": 3840, "last_word_output_sample": 1009920, "samples": 1017120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u03-c05.mp3" }, "en-conversation-05-podcast-block-free-u03-c06": { "id": "en-conversation-05-podcast-block-free-u03-c06", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 77655, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 21.36, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1017600, "script_start_s": 0.16, "script_end_s": 21.2, "all_asr_last_word_s": 21.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u03-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.5854415912763352, "source_start_sample": 3840, "source_end_sample": 1024800, "first_word_output_sample": 3840, "last_word_output_sample": 1013760, "samples": 1020960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u03-c06.mp3" }, "en-conversation-05-podcast-block-free-u03-c07": { "id": "en-conversation-05-podcast-block-free-u03-c07", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 77656, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 21.6, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1029120, "script_start_s": 0.16, "script_end_s": 21.44, "all_asr_last_word_s": 21.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u03-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.028985507246376812, "source_identity_reward": 0.5689961880121771, "source_start_sample": 3840, "source_end_sample": 1036320, "first_word_output_sample": 3840, "last_word_output_sample": 1025280, "samples": 1032480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u03-c07.mp3" }, "de-conversation-05-podcast-block-free-u03-c00": { "id": "de-conversation-05-podcast-block-free-u03-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 87649, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 29.68, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1128960, "script_start_s": 0.0, "script_end_s": 23.52, "all_asr_last_word_s": 29.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3142857142857143, "source_identity_reward": 0.6479895038298387, "source_start_sample": 0, "source_end_sample": 1136160, "first_word_output_sample": 0, "last_word_output_sample": 1128960, "samples": 1136160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 6.01, "added_silence_s": 0.0, "duration_s": 23.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u03-c00.mp3" }, "de-conversation-05-podcast-block-free-u03-c01": { "id": "de-conversation-05-podcast-block-free-u03-c01", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 87650, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 29.92, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1132800, "script_start_s": 0.0, "script_end_s": 23.6, "all_asr_last_word_s": 28.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u03-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.14285714285714285, "source_identity_reward": 0.7071593866032334, "source_start_sample": 0, "source_end_sample": 1140000, "first_word_output_sample": 0, "last_word_output_sample": 1132800, "samples": 1140000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 6.17, "added_silence_s": 0.0, "duration_s": 23.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u03-c01.mp3" }, "de-conversation-05-podcast-block-free-u03-c02": { "id": "de-conversation-05-podcast-block-free-u03-c02", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 87651, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.32, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1159680, "script_start_s": 0.08, "script_end_s": 24.16, "all_asr_last_word_s": 24.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u03-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04285714285714286, "source_identity_reward": 0.6548534859658146, "source_start_sample": 0, "source_end_sample": 1166880, "first_word_output_sample": 3840, "last_word_output_sample": 1159680, "samples": 1166880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 24.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u03-c02.mp3" }, "de-conversation-05-podcast-block-free-u03-c03": { "id": "de-conversation-05-podcast-block-free-u03-c03", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 87652, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 29.2, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 68, "coverage": 0.9714285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1171200, "script_start_s": 0.0, "script_end_s": 24.4, "all_asr_last_word_s": 28.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u03-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.21428571428571427, "source_identity_reward": 0.6965665194035234, "source_start_sample": 0, "source_end_sample": 1178400, "first_word_output_sample": 0, "last_word_output_sample": 1171200, "samples": 1178400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.65, "added_silence_s": 0.0, "duration_s": 24.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u03-c03.mp3" }, "de-conversation-05-podcast-block-free-u03-c04": { "id": "de-conversation-05-podcast-block-free-u03-c04", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 87653, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 23.52, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 68, "coverage": 0.9714285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1121280, "script_start_s": 0.16, "script_end_s": 23.36, "all_asr_last_word_s": 23.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u03-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05714285714285714, "source_identity_reward": 0.800061161892196, "source_start_sample": 3840, "source_end_sample": 1128480, "first_word_output_sample": 3840, "last_word_output_sample": 1117440, "samples": 1124640, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u03-c04.mp3" }, "de-conversation-05-podcast-block-free-u03-c05": { "id": "de-conversation-05-podcast-block-free-u03-c05", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 87654, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.48, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 67, "coverage": 0.9571428571428572, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1167360, "script_start_s": 0.0, "script_end_s": 24.32, "all_asr_last_word_s": 24.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u03-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07142857142857142, "source_identity_reward": 0.8047522276079351, "source_start_sample": 0, "source_end_sample": 1174560, "first_word_output_sample": 0, "last_word_output_sample": 1167360, "samples": 1174560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 24.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u03-c05.mp3" }, "de-conversation-05-podcast-block-free-u03-c06": { "id": "de-conversation-05-podcast-block-free-u03-c06", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 87655, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.16, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1152000, "script_start_s": 0.0, "script_end_s": 24.0, "all_asr_last_word_s": 24.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u03-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04285714285714286, "source_identity_reward": 0.7605239335619356, "source_start_sample": 0, "source_end_sample": 1159200, "first_word_output_sample": 0, "last_word_output_sample": 1152000, "samples": 1159200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 24.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u03-c06.mp3" }, "de-conversation-05-podcast-block-free-u03-c07": { "id": "de-conversation-05-podcast-block-free-u03-c07", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 87656, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 23.92, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 68, "coverage": 0.9714285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1140480, "script_start_s": 0.0, "script_end_s": 23.76, "all_asr_last_word_s": 23.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u03-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07142857142857142, "source_identity_reward": 0.817972584292048, "source_start_sample": 0, "source_end_sample": 1147680, "first_word_output_sample": 0, "last_word_output_sample": 1140480, "samples": 1147680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u03-c07.mp3" }, "en-conversation-01-podcast-segment-free-u10-c00": { "id": "en-conversation-01-podcast-segment-free-u10-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 276480, "script_start_s": 0.16, "script_end_s": 5.76, "all_asr_last_word_s": 29.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.5, "source_identity_reward": 0.183793245494972, "source_start_sample": 3840, "source_end_sample": 283680, "first_word_output_sample": 3840, "last_word_output_sample": 272640, "samples": 279840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 24.09, "added_silence_s": 0.0, "duration_s": 5.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-free-u10-c00.mp3" }, "de-conversation-04-podcast-segment-free-u01-c00": { "id": "de-conversation-04-podcast-segment-free-u01-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86117, "plain": "Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde?", "source_duration_s": 14.56, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 264960, "script_start_s": 0.08, "script_end_s": 5.52, "all_asr_last_word_s": 7.6000000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.18181818181818182, "source_identity_reward": 0.4698053882941285, "source_start_sample": 0, "source_end_sample": 272160, "first_word_output_sample": 3840, "last_word_output_sample": 264960, "samples": 272160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 8.89, "added_silence_s": 0.0, "duration_s": 5.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-free-u01-c00.mp3" }, "de-conversation-01-podcast-segment-timed-u10-c00": { "id": "de-conversation-01-podcast-segment-timed-u10-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 5.68, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 268800, "script_start_s": 0.0, "script_end_s": 5.6, "all_asr_last_word_s": 5.6000000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7318034615456959, "source_start_sample": 0, "source_end_sample": 272640, "first_word_output_sample": 0, "last_word_output_sample": 268800, "samples": 276000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-timed-u10-c00.mp3" }, "en-conversation-05-podcast-segment-timed-u14-c00": { "id": "en-conversation-05-podcast-segment-timed-u14-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77325, "plain": "Exactly. The exchange should feel responsive, with the same two people returning after every change of mood.", "source_duration_s": 5.68, "source_frame_cap": false, "alignment": { "expected_tokens": 17, "matched_tokens": 17, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 268800, "script_start_s": 0.16, "script_end_s": 5.6, "all_asr_last_word_s": 5.6000000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u14-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5236628936028093, "source_start_sample": 3840, "source_end_sample": 272640, "first_word_output_sample": 3840, "last_word_output_sample": 264960, "samples": 272160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-timed-u14-c00.mp3" }, "en-conversation-04-podcast-segment-free-u00-c00": { "id": "en-conversation-04-podcast-segment-free-u00-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76101, "plain": "I can write a requested length next to an individual sentence in the script.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 234240, "script_start_s": 0.0, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5534535806966209, "source_start_sample": 0, "source_end_sample": 241440, "first_word_output_sample": 0, "last_word_output_sample": 234240, "samples": 241440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-free-u00-c00.mp3" }, "en-conversation-04-podcast-segment-free-u08-c00": { "id": "en-conversation-04-podcast-segment-free-u08-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 10.56, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 261120, "script_start_s": 0.0, "script_end_s": 5.44, "all_asr_last_word_s": 5.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3687133849374782, "source_start_sample": 0, "source_end_sample": 268320, "first_word_output_sample": 0, "last_word_output_sample": 261120, "samples": 268320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.97, "added_silence_s": 0.0, "duration_s": 5.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-free-u08-c00.mp3" }, "de-conversation-05-podcast-segment-timed-u11-c00": { "id": "de-conversation-05-podcast-segment-timed-u11-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 222720, "script_start_s": 0.0, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5983736308953063, "source_start_sample": 0, "source_end_sample": 226560, "first_word_output_sample": 0, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-timed-u11-c00.mp3" }, "en-conversation-02-podcast-segment-free-u09-c00": { "id": "en-conversation-02-podcast-segment-free-u09-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 7.92, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 376320, "script_start_s": 0.16, "script_end_s": 7.84, "all_asr_last_word_s": 7.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.14448383574271895, "source_start_sample": 3840, "source_end_sample": 380160, "first_word_output_sample": 3840, "last_word_output_sample": 372480, "samples": 379680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 7.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-free-u09-c00.mp3" }, "de-conversation-04-podcast-block-timed-u02-c00": { "id": "de-conversation-04-podcast-block-timed-u02-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 86633, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 487680, "script_start_s": 0.0, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6901188833905306, "source_start_sample": 0, "source_end_sample": 494880, "first_word_output_sample": 0, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u02-c00.mp3" }, "de-conversation-04-podcast-block-timed-u02-c01": { "id": "de-conversation-04-podcast-block-timed-u02-c01", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 86634, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 487680, "script_start_s": 0.0, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7832284074760597, "source_start_sample": 0, "source_end_sample": 494880, "first_word_output_sample": 0, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u02-c01.mp3" }, "de-conversation-04-podcast-block-timed-u02-c02": { "id": "de-conversation-04-podcast-block-timed-u02-c02", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 86635, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 487680, "script_start_s": 0.08, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7744869321617552, "source_start_sample": 0, "source_end_sample": 494880, "first_word_output_sample": 3840, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u02-c02.mp3" }, "de-conversation-04-podcast-block-timed-u02-c03": { "id": "de-conversation-04-podcast-block-timed-u02-c03", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 86636, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 487680, "script_start_s": 0.0, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7565767705111361, "source_start_sample": 0, "source_end_sample": 494880, "first_word_output_sample": 0, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u02-c03.mp3" }, "de-conversation-04-podcast-block-timed-u02-c04": { "id": "de-conversation-04-podcast-block-timed-u02-c04", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 86637, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 491520, "script_start_s": 0.0, "script_end_s": 10.24, "all_asr_last_word_s": 10.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7199910161571457, "source_start_sample": 0, "source_end_sample": 495360, "first_word_output_sample": 0, "last_word_output_sample": 491520, "samples": 498720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 10.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u02-c04.mp3" }, "de-conversation-04-podcast-block-timed-u02-c05": { "id": "de-conversation-04-podcast-block-timed-u02-c05", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 86638, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 487680, "script_start_s": 0.0, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7587828414015326, "source_start_sample": 0, "source_end_sample": 494880, "first_word_output_sample": 0, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u02-c05.mp3" }, "de-conversation-04-podcast-block-timed-u02-c06": { "id": "de-conversation-04-podcast-block-timed-u02-c06", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 86639, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 487680, "script_start_s": 0.0, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6390806609436651, "source_start_sample": 0, "source_end_sample": 494880, "first_word_output_sample": 0, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u02-c06.mp3" }, "de-conversation-04-podcast-block-timed-u02-c07": { "id": "de-conversation-04-podcast-block-timed-u02-c07", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 86640, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 491520, "script_start_s": 0.0, "script_end_s": 10.24, "all_asr_last_word_s": 10.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.717563452357933, "source_start_sample": 0, "source_end_sample": 495360, "first_word_output_sample": 0, "last_word_output_sample": 491520, "samples": 498720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 10.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u02-c07.mp3" }, "en-conversation-02-podcast-segment-free-u05-c00": { "id": "en-conversation-02-podcast-segment-free-u05-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74181, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change.", "source_duration_s": 16.56, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 549120, "script_start_s": 0.24, "script_end_s": 11.44, "all_asr_last_word_s": 16.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.8, "source_identity_reward": 0.09469279606207612, "source_start_sample": 7680, "source_end_sample": 556320, "first_word_output_sample": 3840, "last_word_output_sample": 541440, "samples": 548640, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 4.97, "added_silence_s": 0.0, "duration_s": 11.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-free-u05-c00.mp3" }, "en-conversation-05-podcast-segment-free-u07-c00": { "id": "en-conversation-05-podcast-segment-free-u07-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77213, "plain": "All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 16, "matched_tokens": 14, "coverage": 0.875, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 2.625, "source_identity_reward": 0.038370539747202566 }, "en-conversation-03-podcast-block-free-u01-c00": { "id": "en-conversation-03-podcast-block-free-u01-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 75617, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 18.08, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 860160, "script_start_s": 0.16, "script_end_s": 17.92, "all_asr_last_word_s": 17.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5802970992133167, "source_start_sample": 3840, "source_end_sample": 867360, "first_word_output_sample": 3840, "last_word_output_sample": 856320, "samples": 863520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u01-c00.mp3" }, "en-conversation-03-podcast-block-free-u01-c01": { "id": "en-conversation-03-podcast-block-free-u01-c01", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 75618, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 18.48, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 50, "coverage": 0.9803921568627451, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0196078431372549, "source_identity_reward": 0.47514614262442123 }, "en-conversation-03-podcast-block-free-u01-c02": { "id": "en-conversation-03-podcast-block-free-u01-c02", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 75619, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 20.16, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 50, "coverage": 0.9803921568627451, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0196078431372549, "source_identity_reward": 0.5167308124428587 }, "en-conversation-03-podcast-block-free-u01-c03": { "id": "en-conversation-03-podcast-block-free-u01-c03", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 75620, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 18.16, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 40, "coverage": 0.7843137254901961, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.23529411764705882, "source_identity_reward": 0.26963920549816034 }, "en-conversation-03-podcast-block-free-u01-c04": { "id": "en-conversation-03-podcast-block-free-u01-c04", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 75621, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 871680, "script_start_s": 0.16, "script_end_s": 18.16, "all_asr_last_word_s": 18.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.47238679714143783, "source_start_sample": 3840, "source_end_sample": 878880, "first_word_output_sample": 3840, "last_word_output_sample": 867840, "samples": 875040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 18.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u01-c04.mp3" }, "en-conversation-03-podcast-block-free-u01-c05": { "id": "en-conversation-03-podcast-block-free-u01-c05", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 75622, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 18.16, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 50, "coverage": 0.9803921568627451, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 864000, "script_start_s": 0.08, "script_end_s": 18.0, "all_asr_last_word_s": 18.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0196078431372549, "source_identity_reward": 0.4798692205681485, "source_start_sample": 0, "source_end_sample": 871200, "first_word_output_sample": 3840, "last_word_output_sample": 864000, "samples": 871200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 18.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u01-c05.mp3" }, "en-conversation-03-podcast-block-free-u01-c06": { "id": "en-conversation-03-podcast-block-free-u01-c06", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 75623, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 20.64, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 890880, "script_start_s": 0.0, "script_end_s": 18.56, "all_asr_last_word_s": 18.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.549706001745647, "source_start_sample": 0, "source_end_sample": 898080, "first_word_output_sample": 0, "last_word_output_sample": 890880, "samples": 898080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.93, "added_silence_s": 0.0, "duration_s": 18.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u01-c06.mp3" }, "en-conversation-03-podcast-block-free-u01-c07": { "id": "en-conversation-03-podcast-block-free-u01-c07", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 75624, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 22.48, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 902400, "script_start_s": 0.0, "script_end_s": 18.8, "all_asr_last_word_s": 18.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.46102018460317035, "source_start_sample": 0, "source_end_sample": 909600, "first_word_output_sample": 0, "last_word_output_sample": 902400, "samples": 909600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 3.53, "added_silence_s": 0.0, "duration_s": 18.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u01-c07.mp3" }, "de-conversation-01-podcast-segment-timed-u03-c00": { "id": "de-conversation-01-podcast-segment-timed-u03-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83149, "plain": "Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 4.0, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 10, "coverage": 0.9090909090909091, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 188160, "script_start_s": 0.16, "script_end_s": 3.92, "all_asr_last_word_s": 3.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09090909090909091, "source_identity_reward": 0.47144379977457046, "source_start_sample": 3840, "source_end_sample": 192000, "first_word_output_sample": 3840, "last_word_output_sample": 184320, "samples": 191520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-timed-u03-c00.mp3" }, "de-conversation-02-podcast-segment-timed-u09-c00": { "id": "de-conversation-02-podcast-segment-timed-u09-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 234240, "script_start_s": 0.0, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.38725052435213414, "source_start_sample": 0, "source_end_sample": 241440, "first_word_output_sample": 0, "last_word_output_sample": 234240, "samples": 241440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-timed-u09-c00.mp3" }, "en-conversation-02-podcast-block-free-u00-c00": { "id": "en-conversation-02-podcast-block-free-u00-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 74601, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 23.12, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 67, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1102080, "script_start_s": 0.0, "script_end_s": 22.96, "all_asr_last_word_s": 22.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.47337097218421115, "source_start_sample": 0, "source_end_sample": 1109280, "first_word_output_sample": 0, "last_word_output_sample": 1102080, "samples": 1109280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u00-c00.mp3" }, "en-conversation-02-podcast-block-free-u00-c01": { "id": "en-conversation-02-podcast-block-free-u00-c01", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 74602, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 67, "matched_tokens": 67, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1163520, "script_start_s": 0.16, "script_end_s": 24.24, "all_asr_last_word_s": 33.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2537313432835821, "source_identity_reward": 0.35634237870672314, "source_start_sample": 3840, "source_end_sample": 1170720, "first_word_output_sample": 3840, "last_word_output_sample": 1159680, "samples": 1166880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 35.61, "added_silence_s": 0.0, "duration_s": 24.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u00-c01.mp3" }, "en-conversation-02-podcast-block-free-u00-c02": { "id": "en-conversation-02-podcast-block-free-u00-c02", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 74603, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.4, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1067520, "script_start_s": 0.16, "script_end_s": 22.24, "all_asr_last_word_s": 22.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.5079558096379384, "source_start_sample": 3840, "source_end_sample": 1074720, "first_word_output_sample": 3840, "last_word_output_sample": 1063680, "samples": 1070880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u00-c02.mp3" }, "en-conversation-02-podcast-block-free-u00-c03": { "id": "en-conversation-02-podcast-block-free-u00-c03", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 74604, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 23.92, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1140480, "script_start_s": 0.16, "script_end_s": 23.76, "all_asr_last_word_s": 23.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.49885321325348453, "source_start_sample": 3840, "source_end_sample": 1147680, "first_word_output_sample": 3840, "last_word_output_sample": 1136640, "samples": 1143840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u00-c03.mp3" }, "en-conversation-02-podcast-block-free-u00-c04": { "id": "en-conversation-02-podcast-block-free-u00-c04", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 74605, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 24.32, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1155840, "script_start_s": 0.0, "script_end_s": 24.08, "all_asr_last_word_s": 24.080000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.4208180684486723, "source_start_sample": 0, "source_end_sample": 1163040, "first_word_output_sample": 0, "last_word_output_sample": 1155840, "samples": 1163040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 24.23, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u00-c04.mp3" }, "en-conversation-02-podcast-block-free-u00-c05": { "id": "en-conversation-02-podcast-block-free-u00-c05", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 74606, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 26.72, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04477611940298507, "source_identity_reward": 0.5420331282221161 }, "en-conversation-02-podcast-block-free-u00-c06": { "id": "en-conversation-02-podcast-block-free-u00-c06", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 74607, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 55.68, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1159680, "script_start_s": 0.16, "script_end_s": 24.16, "all_asr_last_word_s": 47.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.29850746268656714, "source_identity_reward": 0.34776244142820134, "source_start_sample": 3840, "source_end_sample": 1166880, "first_word_output_sample": 3840, "last_word_output_sample": 1155840, "samples": 1163040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 31.37, "added_silence_s": 0.0, "duration_s": 24.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u00-c06.mp3" }, "en-conversation-02-podcast-block-free-u00-c07": { "id": "en-conversation-02-podcast-block-free-u00-c07", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 74608, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 23.36, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1113600, "script_start_s": 0.16, "script_end_s": 23.2, "all_asr_last_word_s": 23.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.6137279351430637, "source_start_sample": 3840, "source_end_sample": 1120800, "first_word_output_sample": 3840, "last_word_output_sample": 1109760, "samples": 1116960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u00-c07.mp3" }, "de-conversation-01-podcast-segment-free-u05-c00": { "id": "de-conversation-01-podcast-segment-free-u05-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83181, "plain": "Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 318720, "script_start_s": 0.0, "script_end_s": 6.64, "all_asr_last_word_s": 7.5200000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09090909090909091, "source_identity_reward": 0.25307260067960413, "source_start_sample": 0, "source_end_sample": 325920, "first_word_output_sample": 0, "last_word_output_sample": 318720, "samples": 325920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.21, "added_silence_s": 0.0, "duration_s": 6.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-free-u05-c00.mp3" }, "en-conversation-03-podcast-segment-free-u07-c00": { "id": "en-conversation-03-podcast-segment-free-u07-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75213, "plain": "Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 7.2, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 238080, "script_start_s": 0.0, "script_end_s": 4.96, "all_asr_last_word_s": 4.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4318626992217335, "source_start_sample": 0, "source_end_sample": 245280, "first_word_output_sample": 0, "last_word_output_sample": 238080, "samples": 245280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.09, "added_silence_s": 0.0, "duration_s": 5.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-free-u07-c00.mp3" }, "en-conversation-01-podcast-segment-free-u05-c00": { "id": "en-conversation-01-podcast-segment-free-u05-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73181, "plain": "That distinction makes me trust the demonstration a little more. Annoyingly sensible of you.", "source_duration_s": 9.84, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 245760, "script_start_s": 0.16, "script_end_s": 5.12, "all_asr_last_word_s": 9.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2857142857142857, "source_identity_reward": 0.08483082939916299, "source_start_sample": 3840, "source_end_sample": 252960, "first_word_output_sample": 3840, "last_word_output_sample": 241920, "samples": 249120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 4.57, "added_silence_s": 0.0, "duration_s": 5.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-free-u05-c00.mp3" }, "en-conversation-01-podcast-block-free-u01-c00": { "id": "en-conversation-01-podcast-block-free-u01-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 73617, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 25.04, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 963840, "script_start_s": 0.0, "script_end_s": 20.08, "all_asr_last_word_s": 24.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.16981132075471697, "source_identity_reward": 0.4153819598910506, "source_start_sample": 0, "source_end_sample": 971040, "first_word_output_sample": 0, "last_word_output_sample": 963840, "samples": 971040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.81, "added_silence_s": 0.0, "duration_s": 20.23, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u01-c00.mp3" }, "en-conversation-01-podcast-block-free-u01-c01": { "id": "en-conversation-01-podcast-block-free-u01-c01", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 73618, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 38.08, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 952320, "script_start_s": 0.0, "script_end_s": 19.84, "all_asr_last_word_s": 37.879999999999995 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.7358490566037735, "source_identity_reward": 0.20625993790068983, "source_start_sample": 0, "source_end_sample": 959520, "first_word_output_sample": 0, "last_word_output_sample": 952320, "samples": 959520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 18.09, "added_silence_s": 0.0, "duration_s": 19.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u01-c01.mp3" }, "en-conversation-01-podcast-block-free-u01-c02": { "id": "en-conversation-01-podcast-block-free-u01-c02", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 73619, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 933120, "script_start_s": 0.16, "script_end_s": 19.44, "all_asr_last_word_s": 19.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.4672300951299706, "source_start_sample": 3840, "source_end_sample": 940320, "first_word_output_sample": 3840, "last_word_output_sample": 929280, "samples": 936480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 19.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u01-c02.mp3" }, "en-conversation-01-podcast-block-free-u01-c03": { "id": "en-conversation-01-podcast-block-free-u01-c03", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 73620, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 20.64, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 936960, "script_start_s": 0.0, "script_end_s": 19.52, "all_asr_last_word_s": 19.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.3811980439564235, "source_start_sample": 0, "source_end_sample": 944160, "first_word_output_sample": 0, "last_word_output_sample": 936960, "samples": 944160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.97, "added_silence_s": 0.0, "duration_s": 19.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u01-c03.mp3" }, "en-conversation-01-podcast-block-free-u01-c04": { "id": "en-conversation-01-podcast-block-free-u01-c04", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 73621, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 23.36, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 48, "coverage": 0.9056603773584906, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 913920, "script_start_s": 0.0, "script_end_s": 19.04, "all_asr_last_word_s": 21.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.16981132075471697, "source_identity_reward": 0.4603916609990541, "source_start_sample": 0, "source_end_sample": 921120, "first_word_output_sample": 0, "last_word_output_sample": 913920, "samples": 921120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.17, "added_silence_s": 0.0, "duration_s": 19.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u01-c04.mp3" }, "en-conversation-01-podcast-block-free-u01-c05": { "id": "en-conversation-01-podcast-block-free-u01-c05", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 73622, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 933120, "script_start_s": 0.0, "script_end_s": 19.44, "all_asr_last_word_s": 19.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.46880192854376135, "source_start_sample": 0, "source_end_sample": 940320, "first_word_output_sample": 0, "last_word_output_sample": 933120, "samples": 940320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 19.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u01-c05.mp3" }, "en-conversation-01-podcast-block-free-u01-c06": { "id": "en-conversation-01-podcast-block-free-u01-c06", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 73623, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 19.44, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 921600, "script_start_s": 0.0, "script_end_s": 19.2, "all_asr_last_word_s": 19.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.45575997632411563, "source_start_sample": 0, "source_end_sample": 928800, "first_word_output_sample": 0, "last_word_output_sample": 921600, "samples": 928800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 19.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u01-c06.mp3" }, "en-conversation-01-podcast-block-free-u01-c07": { "id": "en-conversation-01-podcast-block-free-u01-c07", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 73624, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 933120, "script_start_s": 0.0, "script_end_s": 19.44, "all_asr_last_word_s": 19.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.3588797207223246, "source_start_sample": 0, "source_end_sample": 940320, "first_word_output_sample": 0, "last_word_output_sample": 933120, "samples": 940320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 19.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u01-c07.mp3" }, "en-conversation-02-podcast-block-timed-u01-c00": { "id": "en-conversation-02-podcast-block-timed-u01-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 74617, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.28, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1113600, "script_start_s": 0.16, "script_end_s": 23.2, "all_asr_last_word_s": 23.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5056183799855157, "source_start_sample": 3840, "source_end_sample": 1117440, "first_word_output_sample": 3840, "last_word_output_sample": 1109760, "samples": 1116960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u01-c00.mp3" }, "en-conversation-02-podcast-block-timed-u01-c01": { "id": "en-conversation-02-podcast-block-timed-u01-c01", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 74618, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.36, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1117440, "script_start_s": 0.0, "script_end_s": 23.28, "all_asr_last_word_s": 23.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4955179299869547, "source_start_sample": 0, "source_end_sample": 1121280, "first_word_output_sample": 0, "last_word_output_sample": 1117440, "samples": 1124640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u01-c01.mp3" }, "en-conversation-02-podcast-block-timed-u01-c02": { "id": "en-conversation-02-podcast-block-timed-u01-c02", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 74619, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.28, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1113600, "script_start_s": 0.0, "script_end_s": 23.2, "all_asr_last_word_s": 23.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5504632428927012, "source_start_sample": 0, "source_end_sample": 1117440, "first_word_output_sample": 0, "last_word_output_sample": 1113600, "samples": 1120800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u01-c02.mp3" }, "en-conversation-02-podcast-block-timed-u01-c03": { "id": "en-conversation-02-podcast-block-timed-u01-c03", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 74620, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.28, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1113600, "script_start_s": 0.16, "script_end_s": 23.2, "all_asr_last_word_s": 23.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.45962761192691004, "source_start_sample": 3840, "source_end_sample": 1117440, "first_word_output_sample": 3840, "last_word_output_sample": 1109760, "samples": 1116960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u01-c03.mp3" }, "en-conversation-02-podcast-block-timed-u01-c04": { "id": "en-conversation-02-podcast-block-timed-u01-c04", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 74621, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.36, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1117440, "script_start_s": 0.16, "script_end_s": 23.28, "all_asr_last_word_s": 23.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02857142857142857, "source_identity_reward": 0.4492132502090291, "source_start_sample": 3840, "source_end_sample": 1121280, "first_word_output_sample": 3840, "last_word_output_sample": 1113600, "samples": 1120800, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u01-c04.mp3" }, "en-conversation-02-podcast-block-timed-u01-c05": { "id": "en-conversation-02-podcast-block-timed-u01-c05", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 74622, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.28, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014285714285714285, "source_identity_reward": 0.4658608754872993 }, "en-conversation-02-podcast-block-timed-u01-c06": { "id": "en-conversation-02-podcast-block-timed-u01-c06", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 74623, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.28, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 70, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1113600, "script_start_s": 0.16, "script_end_s": 23.2, "all_asr_last_word_s": 23.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5100053201005219, "source_start_sample": 3840, "source_end_sample": 1117440, "first_word_output_sample": 3840, "last_word_output_sample": 1109760, "samples": 1116960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u01-c06.mp3" }, "en-conversation-02-podcast-block-timed-u01-c07": { "id": "en-conversation-02-podcast-block-timed-u01-c07", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 74624, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change. That is why the reference clips sit beside the output. You can compare for yourself. Now that is a demonstration I can enjoy. Keep the little red light on. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.28, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1113600, "script_start_s": 0.16, "script_end_s": 23.2, "all_asr_last_word_s": 23.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014285714285714285, "source_identity_reward": 0.5876025361366083, "source_start_sample": 3840, "source_end_sample": 1117440, "first_word_output_sample": 3840, "last_word_output_sample": 1109760, "samples": 1116960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-timed-u01-c07.mp3" }, "de-conversation-01-podcast-segment-timed-u07-c00": { "id": "de-conversation-01-podcast-segment-timed-u07-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83213, "plain": "Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 13, "coverage": 0.9285714285714286, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 234240, "script_start_s": 0.24, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07142857142857142, "source_identity_reward": 0.6150114752249379, "source_start_sample": 7680, "source_end_sample": 241440, "first_word_output_sample": 3840, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-timed-u07-c00.mp3" }, "de-conversation-03-podcast-block-timed-u01-c00": { "id": "de-conversation-03-podcast-block-timed-u01-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 85617, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.56, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 42, "coverage": 0.8936170212765957, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.10638297872340426, "source_identity_reward": 0.5303868374967966 }, "de-conversation-03-podcast-block-timed-u01-c01": { "id": "de-conversation-03-podcast-block-timed-u01-c01", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 85618, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.8, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 45, "coverage": 0.9574468085106383, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 802560, "script_start_s": 0.0, "script_end_s": 16.72, "all_asr_last_word_s": 16.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0425531914893617, "source_identity_reward": 0.6179868639951501, "source_start_sample": 0, "source_end_sample": 806400, "first_word_output_sample": 0, "last_word_output_sample": 802560, "samples": 809760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u01-c01.mp3" }, "de-conversation-03-podcast-block-timed-u01-c02": { "id": "de-conversation-03-podcast-block-timed-u01-c02", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 85619, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.64, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 44, "coverage": 0.9361702127659575, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06382978723404255, "source_identity_reward": 0.5580460568306042 }, "de-conversation-03-podcast-block-timed-u01-c03": { "id": "de-conversation-03-podcast-block-timed-u01-c03", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 85620, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.72, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 44, "coverage": 0.9361702127659575, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 798720, "script_start_s": 0.0, "script_end_s": 16.64, "all_asr_last_word_s": 16.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06382978723404255, "source_identity_reward": 0.6351734886520507, "source_start_sample": 0, "source_end_sample": 802560, "first_word_output_sample": 0, "last_word_output_sample": 798720, "samples": 805920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u01-c03.mp3" }, "de-conversation-03-podcast-block-timed-u01-c04": { "id": "de-conversation-03-podcast-block-timed-u01-c04", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 85621, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.72, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 46, "coverage": 0.9787234042553191, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 798720, "script_start_s": 0.0, "script_end_s": 16.64, "all_asr_last_word_s": 16.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02127659574468085, "source_identity_reward": 0.6113763570279911, "source_start_sample": 0, "source_end_sample": 802560, "first_word_output_sample": 0, "last_word_output_sample": 798720, "samples": 805920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u01-c04.mp3" }, "de-conversation-03-podcast-block-timed-u01-c05": { "id": "de-conversation-03-podcast-block-timed-u01-c05", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 85622, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.72, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 45, "coverage": 0.9574468085106383, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0425531914893617, "source_identity_reward": 0.6145649313701113 }, "de-conversation-03-podcast-block-timed-u01-c06": { "id": "de-conversation-03-podcast-block-timed-u01-c06", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 85623, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.72, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 45, "coverage": 0.9574468085106383, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0425531914893617, "source_identity_reward": 0.3803240090542162 }, "de-conversation-03-podcast-block-timed-u01-c07": { "id": "de-conversation-03-podcast-block-timed-u01-c07", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 85624, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.72, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 46, "coverage": 0.9787234042553191, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 798720, "script_start_s": 0.0, "script_end_s": 16.64, "all_asr_last_word_s": 16.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02127659574468085, "source_identity_reward": 0.651556182494881, "source_start_sample": 0, "source_end_sample": 802560, "first_word_output_sample": 0, "last_word_output_sample": 798720, "samples": 805920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u01-c07.mp3" }, "en-conversation-02-podcast-segment-timed-u08-c00": { "id": "en-conversation-02-podcast-segment-timed-u08-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 222720, "script_start_s": 0.16, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.20666498017021176, "source_start_sample": 3840, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-timed-u08-c00.mp3" }, "de-conversation-02-podcast-segment-timed-u10-c00": { "id": "de-conversation-02-podcast-segment-timed-u10-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 5.68, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 268800, "script_start_s": 0.0, "script_end_s": 5.6, "all_asr_last_word_s": 5.6000000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8272284555175, "source_start_sample": 0, "source_end_sample": 272640, "first_word_output_sample": 0, "last_word_output_sample": 268800, "samples": 276000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-timed-u10-c00.mp3" }, "de-conversation-01-podcast-block-timed-u02-c00": { "id": "de-conversation-01-podcast-block-timed-u02-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 83633, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 53, "coverage": 0.9464285714285714, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05357142857142857, "source_identity_reward": 0.743635409817585 }, "de-conversation-01-podcast-block-timed-u02-c01": { "id": "de-conversation-01-podcast-block-timed-u02-c01", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 83634, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 952320, "script_start_s": 0.0, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7175807866927912, "source_start_sample": 0, "source_end_sample": 959520, "first_word_output_sample": 0, "last_word_output_sample": 952320, "samples": 959520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u02-c01.mp3" }, "de-conversation-01-podcast-block-timed-u02-c02": { "id": "de-conversation-01-podcast-block-timed-u02-c02", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 83635, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 952320, "script_start_s": 0.0, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7998853283836804, "source_start_sample": 0, "source_end_sample": 959520, "first_word_output_sample": 0, "last_word_output_sample": 952320, "samples": 959520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u02-c02.mp3" }, "de-conversation-01-podcast-block-timed-u02-c03": { "id": "de-conversation-01-podcast-block-timed-u02-c03", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 83636, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8292460900612326, "source_start_sample": 0, "source_end_sample": 960000, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u02-c03.mp3" }, "de-conversation-01-podcast-block-timed-u02-c04": { "id": "de-conversation-01-podcast-block-timed-u02-c04", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 83637, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.08, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7419562943447857, "source_start_sample": 0, "source_end_sample": 963360, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u02-c04.mp3" }, "de-conversation-01-podcast-block-timed-u02-c05": { "id": "de-conversation-01-podcast-block-timed-u02-c05", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 83638, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 952320, "script_start_s": 0.0, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8494923477471193, "source_start_sample": 0, "source_end_sample": 959520, "first_word_output_sample": 0, "last_word_output_sample": 952320, "samples": 959520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u02-c05.mp3" }, "de-conversation-01-podcast-block-timed-u02-c06": { "id": "de-conversation-01-podcast-block-timed-u02-c06", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 83639, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.08, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8023075474574791, "source_start_sample": 0, "source_end_sample": 963360, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u02-c06.mp3" }, "de-conversation-01-podcast-block-timed-u02-c07": { "id": "de-conversation-01-podcast-block-timed-u02-c07", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 83640, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.08, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7709340463389229, "source_start_sample": 0, "source_end_sample": 963360, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u02-c07.mp3" }, "de-conversation-02-podcast-segment-free-u07-c00": { "id": "de-conversation-02-podcast-segment-free-u07-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84213, "plain": "Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an.", "source_duration_s": 18.08, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 215040, "script_start_s": 0.16, "script_end_s": 4.48, "all_asr_last_word_s": 17.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0, "source_identity_reward": 0.3537740057186639, "source_start_sample": 3840, "source_end_sample": 222240, "first_word_output_sample": 3840, "last_word_output_sample": 211200, "samples": 218400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 13.45, "added_silence_s": 0.0, "duration_s": 4.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-free-u07-c00.mp3" }, "en-conversation-05-podcast-segment-free-u04-c00": { "id": "en-conversation-05-podcast-segment-free-u04-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77165, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers.", "source_duration_s": 12.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 284160, "script_start_s": 0.16, "script_end_s": 5.92, "all_asr_last_word_s": 6.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07692307692307693, "source_identity_reward": 0.4443419062988767, "source_start_sample": 3840, "source_end_sample": 291360, "first_word_output_sample": 3840, "last_word_output_sample": 280320, "samples": 287520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 6.57, "added_silence_s": 0.0, "duration_s": 5.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-free-u04-c00.mp3" }, "de-conversation-05-podcast-segment-timed-u15-c00": { "id": "de-conversation-05-podcast-segment-timed-u15-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87341, "plain": "Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 6.8, "source_frame_cap": false, "alignment": { "expected_tokens": 19, "matched_tokens": 19, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 318720, "script_start_s": 0.0, "script_end_s": 6.64, "all_asr_last_word_s": 6.640000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u15-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5898347191777364, "source_start_sample": 0, "source_end_sample": 325920, "first_word_output_sample": 0, "last_word_output_sample": 318720, "samples": 325920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 6.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-timed-u15-c00.mp3" }, "de-conversation-03-podcast-block-timed-u00-c00": { "id": "de-conversation-03-podcast-block-timed-u00-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 85601, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.52, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 47, "coverage": 0.9591836734693877, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 837120, "script_start_s": 0.0, "script_end_s": 17.44, "all_asr_last_word_s": 17.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04081632653061224, "source_identity_reward": 0.8068892184829859, "source_start_sample": 0, "source_end_sample": 840960, "first_word_output_sample": 0, "last_word_output_sample": 837120, "samples": 844320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u00-c00.mp3" }, "de-conversation-03-podcast-block-timed-u00-c01": { "id": "de-conversation-03-podcast-block-timed-u00-c01", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 85602, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.52, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 837120, "script_start_s": 0.0, "script_end_s": 17.44, "all_asr_last_word_s": 17.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7502065771165125, "source_start_sample": 0, "source_end_sample": 840960, "first_word_output_sample": 0, "last_word_output_sample": 837120, "samples": 844320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u00-c01.mp3" }, "de-conversation-03-podcast-block-timed-u00-c02": { "id": "de-conversation-03-podcast-block-timed-u00-c02", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 85603, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.44, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 48, "coverage": 0.9795918367346939, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02040816326530612, "source_identity_reward": 0.7394887015034441 }, "de-conversation-03-podcast-block-timed-u00-c03": { "id": "de-conversation-03-podcast-block-timed-u00-c03", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 85604, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.52, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 837120, "script_start_s": 0.0, "script_end_s": 17.44, "all_asr_last_word_s": 17.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8588959519051358, "source_start_sample": 0, "source_end_sample": 840960, "first_word_output_sample": 0, "last_word_output_sample": 837120, "samples": 844320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u00-c03.mp3" }, "de-conversation-03-podcast-block-timed-u00-c04": { "id": "de-conversation-03-podcast-block-timed-u00-c04", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 85605, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.52, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 837120, "script_start_s": 0.0, "script_end_s": 17.44, "all_asr_last_word_s": 17.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8047815991155405, "source_start_sample": 0, "source_end_sample": 840960, "first_word_output_sample": 0, "last_word_output_sample": 837120, "samples": 844320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u00-c04.mp3" }, "de-conversation-03-podcast-block-timed-u00-c05": { "id": "de-conversation-03-podcast-block-timed-u00-c05", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 85606, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.52, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 837120, "script_start_s": 0.0, "script_end_s": 17.44, "all_asr_last_word_s": 17.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8136447447490892, "source_start_sample": 0, "source_end_sample": 840960, "first_word_output_sample": 0, "last_word_output_sample": 837120, "samples": 844320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u00-c05.mp3" }, "de-conversation-03-podcast-block-timed-u00-c06": { "id": "de-conversation-03-podcast-block-timed-u00-c06", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 85607, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.44, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 833280, "script_start_s": 0.0, "script_end_s": 17.36, "all_asr_last_word_s": 17.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.802973435884796, "source_start_sample": 0, "source_end_sample": 837120, "first_word_output_sample": 0, "last_word_output_sample": 833280, "samples": 840480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u00-c06.mp3" }, "de-conversation-03-podcast-block-timed-u00-c07": { "id": "de-conversation-03-podcast-block-timed-u00-c07", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 85608, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 17.52, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 837120, "script_start_s": 0.0, "script_end_s": 17.44, "all_asr_last_word_s": 17.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8051190734445992, "source_start_sample": 0, "source_end_sample": 840960, "first_word_output_sample": 0, "last_word_output_sample": 837120, "samples": 844320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-timed-u00-c07.mp3" }, "en-conversation-02-podcast-segment-free-u01-c00": { "id": "en-conversation-02-podcast-segment-free-u01-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74117, "plain": "Please tell me the tiny studio has a very dramatic red recording light.", "source_duration_s": 10.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 268800, "script_start_s": 0.16, "script_end_s": 5.6, "all_asr_last_word_s": 10.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.8461538461538461, "source_identity_reward": 0.16298401726605435, "source_start_sample": 3840, "source_end_sample": 276000, "first_word_output_sample": 3840, "last_word_output_sample": 264960, "samples": 272160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 4.97, "added_silence_s": 0.0, "duration_s": 5.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-free-u01-c00.mp3" }, "en-conversation-03-podcast-segment-free-u04-c00": { "id": "en-conversation-03-podcast-segment-free-u04-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75165, "plain": "We test that directly, turn by turn, rather than trusting the first sentence.", "source_duration_s": 7.6, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 238080, "script_start_s": 0.0, "script_end_s": 4.96, "all_asr_last_word_s": 4.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4294480590777191, "source_start_sample": 0, "source_end_sample": 245280, "first_word_output_sample": 0, "last_word_output_sample": 238080, "samples": 245280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.49, "added_silence_s": 0.0, "duration_s": 5.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-free-u04-c00.mp3" }, "en-conversation-01-podcast-segment-timed-u02-c00": { "id": "en-conversation-01-podcast-segment-timed-u02-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73133, "plain": "It can read a script with different speakers and sentence-level acting directions.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 12, "coverage": 0.9230769230769231, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07692307692307693, "source_identity_reward": 0.425381154982261 }, "de-conversation-03-podcast-segment-free-u01-c00": { "id": "de-conversation-03-podcast-segment-free-u01-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85117, "plain": "Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt?", "source_duration_s": 9.36, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 311040, "script_start_s": 0.0, "script_end_s": 6.48, "all_asr_last_word_s": 6.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.661430105938259, "source_start_sample": 0, "source_end_sample": 318240, "first_word_output_sample": 0, "last_word_output_sample": 311040, "samples": 318240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.73, "added_silence_s": 0.0, "duration_s": 6.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-free-u01-c00.mp3" }, "de-conversation-01-podcast-block-free-u00-c00": { "id": "de-conversation-01-podcast-block-free-u00-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 83601, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 55.92, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 42, "coverage": 0.9545454545454546, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 867840, "script_start_s": 0.0, "script_end_s": 18.08, "all_asr_last_word_s": 55.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0227272727272727, "source_identity_reward": 0.3200871062233286, "source_start_sample": 0, "source_end_sample": 875040, "first_word_output_sample": 0, "last_word_output_sample": 867840, "samples": 875040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 37.69, "added_silence_s": 0.0, "duration_s": 18.23, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u00-c00.mp3" }, "de-conversation-01-podcast-block-free-u00-c01": { "id": "de-conversation-01-podcast-block-free-u00-c01", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 83602, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 28.24, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 41, "coverage": 0.9318181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 867840, "script_start_s": 0.0, "script_end_s": 18.08, "all_asr_last_word_s": 27.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5227272727272727, "source_identity_reward": 0.5198276831250563, "source_start_sample": 0, "source_end_sample": 875040, "first_word_output_sample": 0, "last_word_output_sample": 867840, "samples": 875040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 10.01, "added_silence_s": 0.0, "duration_s": 18.23, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u00-c01.mp3" }, "de-conversation-01-podcast-block-free-u00-c02": { "id": "de-conversation-01-podcast-block-free-u00-c02", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 83603, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 20.16, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 42, "coverage": 0.9545454545454546, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 848640, "script_start_s": 0.0, "script_end_s": 17.68, "all_asr_last_word_s": 19.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.11363636363636363, "source_identity_reward": 0.7568893327796273, "source_start_sample": 0, "source_end_sample": 855840, "first_word_output_sample": 0, "last_word_output_sample": 848640, "samples": 855840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.33, "added_silence_s": 0.0, "duration_s": 17.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u00-c02.mp3" }, "de-conversation-01-podcast-block-free-u00-c03": { "id": "de-conversation-01-podcast-block-free-u00-c03", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 83604, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 21.28, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 42, "coverage": 0.9545454545454546, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 879360, "script_start_s": 0.0, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.045454545454545456, "source_identity_reward": 0.7081949036001463, "source_start_sample": 0, "source_end_sample": 886560, "first_word_output_sample": 0, "last_word_output_sample": 879360, "samples": 886560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.81, "added_silence_s": 0.0, "duration_s": 18.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u00-c03.mp3" }, "de-conversation-01-podcast-block-free-u00-c04": { "id": "de-conversation-01-podcast-block-free-u00-c04", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 83605, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 17.68, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 41, "coverage": 0.9318181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 840960, "script_start_s": 0.0, "script_end_s": 17.52, "all_asr_last_word_s": 17.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06818181818181818, "source_identity_reward": 0.7096876756778444, "source_start_sample": 0, "source_end_sample": 848160, "first_word_output_sample": 0, "last_word_output_sample": 840960, "samples": 848160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u00-c04.mp3" }, "de-conversation-01-podcast-block-free-u00-c05": { "id": "de-conversation-01-podcast-block-free-u00-c05", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 83606, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 17.6, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 40, "coverage": 0.9090909090909091, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 833280, "script_start_s": 0.0, "script_end_s": 17.36, "all_asr_last_word_s": 17.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09090909090909091, "source_identity_reward": 0.7700127422177271, "source_start_sample": 0, "source_end_sample": 840480, "first_word_output_sample": 0, "last_word_output_sample": 833280, "samples": 840480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 17.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u00-c05.mp3" }, "de-conversation-01-podcast-block-free-u00-c06": { "id": "de-conversation-01-podcast-block-free-u00-c06", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 83607, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 22.08, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 41, "coverage": 0.9318181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 856320, "script_start_s": 0.0, "script_end_s": 17.84, "all_asr_last_word_s": 21.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3181818181818182, "source_identity_reward": 0.6046392857103327, "source_start_sample": 0, "source_end_sample": 863520, "first_word_output_sample": 0, "last_word_output_sample": 856320, "samples": 863520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.09, "added_silence_s": 0.0, "duration_s": 17.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u00-c06.mp3" }, "de-conversation-01-podcast-block-free-u00-c07": { "id": "de-conversation-01-podcast-block-free-u00-c07", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 83608, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 18.88, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 41, "coverage": 0.9318181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 894720, "script_start_s": 0.0, "script_end_s": 18.64, "all_asr_last_word_s": 18.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06818181818181818, "source_identity_reward": 0.6450020199256139, "source_start_sample": 0, "source_end_sample": 901920, "first_word_output_sample": 0, "last_word_output_sample": 894720, "samples": 901920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 18.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u00-c07.mp3" }, "en-conversation-04-podcast-segment-free-u01-c00": { "id": "en-conversation-04-podcast-segment-free-u01-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76117, "plain": "You are kidding. Do you mean it lands on an exact millisecond?", "source_duration_s": 19.36, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 215040, "script_start_s": 0.16, "script_end_s": 4.48, "all_asr_last_word_s": 19.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.6666666666666666, "source_identity_reward": 0.08996636056533669, "source_start_sample": 3840, "source_end_sample": 222240, "first_word_output_sample": 3840, "last_word_output_sample": 211200, "samples": 218400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 14.73, "added_silence_s": 0.0, "duration_s": 4.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-free-u01-c00.mp3" }, "en-conversation-05-podcast-segment-timed-u00-c00": { "id": "en-conversation-05-podcast-segment-timed-u00-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77101, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch.", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 207360, "script_start_s": 0.08, "script_end_s": 4.32, "all_asr_last_word_s": 4.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.17056892647144262, "source_start_sample": 0, "source_end_sample": 211200, "first_word_output_sample": 3840, "last_word_output_sample": 207360, "samples": 214560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-timed-u00-c00.mp3" }, "de-conversation-01-podcast-segment-free-u02-c00": { "id": "de-conversation-01-podcast-segment-free-u02-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83133, "plain": "Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 284160, "script_start_s": 0.0, "script_end_s": 5.92, "all_asr_last_word_s": 5.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7958923809778957, "source_start_sample": 0, "source_end_sample": 291360, "first_word_output_sample": 0, "last_word_output_sample": 284160, "samples": 291360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.93, "added_silence_s": 0.0, "duration_s": 6.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-free-u02-c00.mp3" }, "de-conversation-05-podcast-segment-timed-u13-c00": { "id": "de-conversation-05-podcast-segment-timed-u13-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87309, "plain": "Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden.", "source_duration_s": 7.04, "source_frame_cap": false, "alignment": { "expected_tokens": 20, "matched_tokens": 19, "coverage": 0.95, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 330240, "script_start_s": 0.16, "script_end_s": 6.88, "all_asr_last_word_s": 6.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u13-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05, "source_identity_reward": 0.5083502655624301, "source_start_sample": 3840, "source_end_sample": 337440, "first_word_output_sample": 3840, "last_word_output_sample": 326400, "samples": 333600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 6.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-timed-u13-c00.mp3" }, "en-conversation-01-podcast-segment-timed-u01-c00": { "id": "en-conversation-01-podcast-segment-timed-u01-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73117, "plain": "You sound suspiciously excited. Did a speech model actually surprise you?", "source_duration_s": 3.76, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 10, "coverage": 0.9090909090909091, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 168960, "script_start_s": 0.08, "script_end_s": 3.52, "all_asr_last_word_s": 3.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09090909090909091, "source_identity_reward": 0.08462684236128268, "source_start_sample": 0, "source_end_sample": 176160, "first_word_output_sample": 3840, "last_word_output_sample": 168960, "samples": 176160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 3.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-timed-u01-c00.mp3" }, "en-conversation-05-podcast-segment-timed-u15-c00": { "id": "en-conversation-05-podcast-segment-timed-u15-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77341, "plain": "Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 6.32, "source_frame_cap": false, "alignment": { "expected_tokens": 19, "matched_tokens": 19, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 295680, "script_start_s": 0.0, "script_end_s": 6.16, "all_asr_last_word_s": 6.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u15-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.14758219798601385, "source_start_sample": 0, "source_end_sample": 302880, "first_word_output_sample": 0, "last_word_output_sample": 295680, "samples": 302880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 6.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-timed-u15-c00.mp3" }, "en-conversation-05-podcast-segment-free-u00-c00": { "id": "en-conversation-05-podcast-segment-free-u00-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77101, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 311040, "script_start_s": 0.24, "script_end_s": 6.48, "all_asr_last_word_s": 28.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 3.769230769230769, "source_identity_reward": 0.0914714795566513, "source_start_sample": 7680, "source_end_sample": 318240, "first_word_output_sample": 3840, "last_word_output_sample": 303360, "samples": 310560, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 23.37, "added_silence_s": 0.0, "duration_s": 6.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-free-u00-c00.mp3" }, "en-conversation-04-podcast-segment-free-u05-c00": { "id": "en-conversation-04-podcast-segment-free-u05-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76181, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words.", "source_duration_s": 10.88, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 514560, "script_start_s": 0.16, "script_end_s": 10.72, "all_asr_last_word_s": 10.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.28705448700838343, "source_start_sample": 3840, "source_end_sample": 521760, "first_word_output_sample": 3840, "last_word_output_sample": 510720, "samples": 517920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-free-u05-c00.mp3" }, "de-conversation-03-podcast-segment-timed-u01-c00": { "id": "de-conversation-03-podcast-segment-timed-u01-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85117, "plain": "Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt?", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6671873186264127, "source_start_sample": 0, "source_end_sample": 222720, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-timed-u01-c00.mp3" }, "en-conversation-05-podcast-segment-timed-u07-c00": { "id": "en-conversation-05-podcast-segment-timed-u07-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77213, "plain": "All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 5.36, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 14, "coverage": 0.875, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.125, "source_identity_reward": 0.23159168296622015 }, "en-conversation-05-podcast-segment-timed-u02-c00": { "id": "en-conversation-05-podcast-segment-timed-u02-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77133, "plain": "That is what this five-language grid explores, with short and longer conversations.", "source_duration_s": 4.0, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 188160, "script_start_s": 0.0, "script_end_s": 3.92, "all_asr_last_word_s": 3.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4828085309467605, "source_start_sample": 0, "source_end_sample": 192000, "first_word_output_sample": 0, "last_word_output_sample": 188160, "samples": 195360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-timed-u02-c00.mp3" }, "en-conversation-05-podcast-segment-timed-u10-c00": { "id": "en-conversation-05-podcast-segment-timed-u10-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 222720, "script_start_s": 0.16, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.35784607038440414, "source_start_sample": 3840, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-timed-u10-c00.mp3" }, "de-conversation-01-podcast-segment-free-u04-c00": { "id": "de-conversation-01-podcast-segment-free-u04-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83165, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt.", "source_duration_s": 15.12, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 11, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 311040, "script_start_s": 0.0, "script_end_s": 6.48, "all_asr_last_word_s": 6.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.7304685728286834, "source_start_sample": 0, "source_end_sample": 318240, "first_word_output_sample": 0, "last_word_output_sample": 311040, "samples": 318240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 8.49, "added_silence_s": 0.0, "duration_s": 6.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-free-u04-c00.mp3" }, "de-conversation-01-podcast-segment-timed-u08-c00": { "id": "de-conversation-01-podcast-segment-timed-u08-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 4.8, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 226560, "script_start_s": 0.0, "script_end_s": 4.72, "all_asr_last_word_s": 4.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7387453310916606, "source_start_sample": 0, "source_end_sample": 230400, "first_word_output_sample": 0, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-timed-u08-c00.mp3" }, "en-conversation-01-podcast-segment-free-u00-c00": { "id": "en-conversation-01-podcast-segment-free-u00-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73101, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large.", "source_duration_s": 9.2, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 12, "coverage": 0.8571428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 303360, "script_start_s": 0.16, "script_end_s": 6.32, "all_asr_last_word_s": 9.120000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.6428571428571429, "source_identity_reward": 0.29976545004983096, "source_start_sample": 3840, "source_end_sample": 310560, "first_word_output_sample": 3840, "last_word_output_sample": 299520, "samples": 306720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 2.73, "added_silence_s": 0.0, "duration_s": 6.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-free-u00-c00.mp3" }, "de-conversation-02-podcast-segment-timed-u11-c00": { "id": "de-conversation-02-podcast-segment-timed-u11-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 222720, "script_start_s": 0.16, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3083839470786538, "source_start_sample": 3840, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-timed-u11-c00.mp3" }, "de-conversation-04-podcast-segment-free-u03-c00": { "id": "de-conversation-04-podcast-segment-free-u03-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86149, "plain": "Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen.", "source_duration_s": 7.6, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 264960, "script_start_s": 0.16, "script_end_s": 5.52, "all_asr_last_word_s": 5.5200000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5756583669498461, "source_start_sample": 3840, "source_end_sample": 272160, "first_word_output_sample": 3840, "last_word_output_sample": 261120, "samples": 268320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.93, "added_silence_s": 0.0, "duration_s": 5.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-free-u03-c00.mp3" }, "de-conversation-02-podcast-segment-free-u06-c00": { "id": "de-conversation-02-podcast-segment-free-u06-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84197, "plain": "Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen.", "source_duration_s": 7.28, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 341760, "script_start_s": 0.0, "script_end_s": 7.12, "all_asr_last_word_s": 7.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8289436200494997, "source_start_sample": 0, "source_end_sample": 348960, "first_word_output_sample": 0, "last_word_output_sample": 341760, "samples": 348960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 7.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-free-u06-c00.mp3" }, "en-conversation-01-podcast-segment-free-u07-c00": { "id": "en-conversation-01-podcast-segment-free-u07-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73213, "plain": "All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 21.6, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 12, "coverage": 0.8571428571428571, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5714285714285714, "source_identity_reward": 0.11379827860219982 }, "de-conversation-04-podcast-segment-free-u06-c00": { "id": "de-conversation-04-podcast-segment-free-u06-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86197, "plain": "Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 303360, "script_start_s": 0.16, "script_end_s": 6.32, "all_asr_last_word_s": 6.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7100857574909913, "source_start_sample": 3840, "source_end_sample": 310560, "first_word_output_sample": 3840, "last_word_output_sample": 299520, "samples": 306720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.53, "added_silence_s": 0.0, "duration_s": 6.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-free-u06-c00.mp3" }, "en-conversation-02-podcast-segment-free-u04-c00": { "id": "en-conversation-02-podcast-segment-free-u04-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74165, "plain": "Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 264960, "script_start_s": 0.08, "script_end_s": 5.52, "all_asr_last_word_s": 8.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2857142857142857, "source_identity_reward": 0.2621902140515044, "source_start_sample": 0, "source_end_sample": 272160, "first_word_output_sample": 3840, "last_word_output_sample": 264960, "samples": 272160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.33, "added_silence_s": 0.0, "duration_s": 5.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-free-u04-c00.mp3" }, "de-conversation-03-podcast-segment-timed-u05-c00": { "id": "de-conversation-03-podcast-segment-timed-u05-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85181, "plain": "Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt.", "source_duration_s": 4.48, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 211200, "script_start_s": 0.16, "script_end_s": 4.4, "all_asr_last_word_s": 4.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4531899078169499, "source_start_sample": 3840, "source_end_sample": 215040, "first_word_output_sample": 3840, "last_word_output_sample": 207360, "samples": 214560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-timed-u05-c00.mp3" }, "de-conversation-03-podcast-segment-timed-u03-c00": { "id": "de-conversation-03-podcast-segment-timed-u03-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85149, "plain": "Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 207360, "script_start_s": 0.0, "script_end_s": 4.32, "all_asr_last_word_s": 4.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.49726424331248537, "source_start_sample": 0, "source_end_sample": 211200, "first_word_output_sample": 0, "last_word_output_sample": 207360, "samples": 214560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-timed-u03-c00.mp3" }, "en-conversation-04-podcast-block-timed-u01-c00": { "id": "en-conversation-04-podcast-block-timed-u01-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 76617, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.88, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 998400, "script_start_s": 0.08, "script_end_s": 20.8, "all_asr_last_word_s": 20.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.44364516103221113, "source_start_sample": 0, "source_end_sample": 1002240, "first_word_output_sample": 3840, "last_word_output_sample": 998400, "samples": 1005600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-timed-u01-c00.mp3" }, "en-conversation-04-podcast-block-timed-u01-c01": { "id": "en-conversation-04-podcast-block-timed-u01-c01", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 76618, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.88, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 998400, "script_start_s": 0.16, "script_end_s": 20.8, "all_asr_last_word_s": 20.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5223057325721259, "source_start_sample": 3840, "source_end_sample": 1002240, "first_word_output_sample": 3840, "last_word_output_sample": 994560, "samples": 1001760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-timed-u01-c01.mp3" }, "en-conversation-04-podcast-block-timed-u01-c02": { "id": "en-conversation-04-podcast-block-timed-u01-c02", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 76619, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.96, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 62, "coverage": 0.9841269841269841, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.015873015873015872, "source_identity_reward": 0.4482161631840584 }, "en-conversation-04-podcast-block-timed-u01-c03": { "id": "en-conversation-04-podcast-block-timed-u01-c03", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 76620, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.88, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 61, "coverage": 0.9682539682539683, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.031746031746031744, "source_identity_reward": 0.5553423941093392 }, "en-conversation-04-podcast-block-timed-u01-c04": { "id": "en-conversation-04-podcast-block-timed-u01-c04", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 76621, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.8, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 61, "coverage": 0.9682539682539683, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.031746031746031744, "source_identity_reward": 0.5564541804847892 }, "en-conversation-04-podcast-block-timed-u01-c05": { "id": "en-conversation-04-podcast-block-timed-u01-c05", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 76622, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.96, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1002240, "script_start_s": 0.16, "script_end_s": 20.88, "all_asr_last_word_s": 20.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4611376692828407, "source_start_sample": 3840, "source_end_sample": 1006080, "first_word_output_sample": 3840, "last_word_output_sample": 998400, "samples": 1005600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-timed-u01-c05.mp3" }, "en-conversation-04-podcast-block-timed-u01-c06": { "id": "en-conversation-04-podcast-block-timed-u01-c06", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 76623, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.88, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 62, "coverage": 0.9841269841269841, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.015873015873015872, "source_identity_reward": 0.5201580432287232 }, "en-conversation-04-podcast-block-timed-u01-c07": { "id": "en-conversation-04-podcast-block-timed-u01-c07", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 76624, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.72, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 60, "coverage": 0.9523809523809523, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.047619047619047616, "source_identity_reward": 0.4904762257375373 }, "de-conversation-04-base17-segment-free-u11-c00": { "id": "de-conversation-04-base17-segment-free-u11-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 403200, "script_start_s": 0.24, "script_end_s": 8.4, "all_asr_last_word_s": 15.040000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.46153846153846156, "source_identity_reward": 0.39162155488271544, "source_start_sample": 7680, "source_end_sample": 410400, "first_word_output_sample": 3840, "last_word_output_sample": 395520, "samples": 402720, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 21.45, "added_silence_s": 0.0, "duration_s": 8.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-free-u11-c00.mp3" }, "de-conversation-04-base17-segment-free-u05-c00": { "id": "de-conversation-04-base17-segment-free-u05-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86181, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 307200, "script_start_s": 0.0, "script_end_s": 6.4, "all_asr_last_word_s": 29.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2, "source_identity_reward": 0.558381778187971, "source_start_sample": 0, "source_end_sample": 314400, "first_word_output_sample": 0, "last_word_output_sample": 307200, "samples": 314400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.45, "added_silence_s": 0.0, "duration_s": 6.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-free-u05-c00.mp3" }, "de-conversation-05-base17-segment-timed-u07-c00": { "id": "de-conversation-05-base17-segment-timed-u07-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87213, "plain": "Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 230400, "script_start_s": 0.16, "script_end_s": 4.8, "all_asr_last_word_s": 4.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5164041913262984, "source_start_sample": 3840, "source_end_sample": 237600, "first_word_output_sample": 3840, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-timed-u07-c00.mp3" }, "de-conversation-02-base17-block-timed-u02-c00": { "id": "de-conversation-02-base17-block-timed-u02-c00", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 84633, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 487680, "script_start_s": 0.16, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.240878264334639, "source_start_sample": 3840, "source_end_sample": 494880, "first_word_output_sample": 3840, "last_word_output_sample": 483840, "samples": 491040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u02-c00.mp3" }, "de-conversation-02-base17-block-timed-u02-c01": { "id": "de-conversation-02-base17-block-timed-u02-c01", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 84634, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 487680, "script_start_s": 0.0, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.41845605014189224, "source_start_sample": 0, "source_end_sample": 494880, "first_word_output_sample": 0, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u02-c01.mp3" }, "de-conversation-02-base17-block-timed-u02-c02": { "id": "de-conversation-02-base17-block-timed-u02-c02", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 84635, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 487680, "script_start_s": 0.08, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.24521042428705597, "source_start_sample": 0, "source_end_sample": 494880, "first_word_output_sample": 3840, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u02-c02.mp3" }, "de-conversation-02-base17-block-timed-u02-c03": { "id": "de-conversation-02-base17-block-timed-u02-c03", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 84636, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 487680, "script_start_s": 0.16, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2519906572209008, "source_start_sample": 3840, "source_end_sample": 494880, "first_word_output_sample": 3840, "last_word_output_sample": 483840, "samples": 491040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u02-c03.mp3" }, "de-conversation-02-base17-block-timed-u02-c04": { "id": "de-conversation-02-base17-block-timed-u02-c04", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 84637, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 491520, "script_start_s": 0.0, "script_end_s": 10.24, "all_asr_last_word_s": 10.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.18562090893477756, "source_start_sample": 0, "source_end_sample": 495360, "first_word_output_sample": 0, "last_word_output_sample": 491520, "samples": 498720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 10.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u02-c04.mp3" }, "de-conversation-02-base17-block-timed-u02-c05": { "id": "de-conversation-02-base17-block-timed-u02-c05", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 84638, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 487680, "script_start_s": 0.16, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.16847475491996367, "source_start_sample": 3840, "source_end_sample": 494880, "first_word_output_sample": 3840, "last_word_output_sample": 483840, "samples": 491040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u02-c05.mp3" }, "de-conversation-02-base17-block-timed-u02-c06": { "id": "de-conversation-02-base17-block-timed-u02-c06", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 84639, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 487680, "script_start_s": 0.16, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.19506672187777785, "source_start_sample": 3840, "source_end_sample": 494880, "first_word_output_sample": 3840, "last_word_output_sample": 483840, "samples": 491040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u02-c06.mp3" }, "de-conversation-02-base17-block-timed-u02-c07": { "id": "de-conversation-02-base17-block-timed-u02-c07", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 84640, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 487680, "script_start_s": 0.16, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.27816957113639146, "source_start_sample": 3840, "source_end_sample": 494880, "first_word_output_sample": 3840, "last_word_output_sample": 483840, "samples": 491040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u02-c07.mp3" }, "de-conversation-05-base17-segment-timed-u03-c00": { "id": "de-conversation-05-base17-segment-timed-u03-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87149, "plain": "Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 11, "coverage": 0.8461538461538461, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.15384615384615385, "source_identity_reward": 0.3952468432705639, "source_start_sample": 0, "source_end_sample": 222720, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-timed-u03-c00.mp3" }, "en-conversation-01-base17-segment-timed-u05-c00": { "id": "en-conversation-01-base17-segment-timed-u05-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73181, "plain": "That distinction makes me trust the demonstration a little more. Annoyingly sensible of you.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 215040, "script_start_s": 0.0, "script_end_s": 4.48, "all_asr_last_word_s": 4.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4094475451861872, "source_start_sample": 0, "source_end_sample": 222240, "first_word_output_sample": 0, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-timed-u05-c00.mp3" }, "de-conversation-02-base17-segment-timed-u06-c00": { "id": "de-conversation-02-base17-segment-timed-u06-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84197, "plain": "Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen.", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 203520, "script_start_s": 0.0, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8191876243461373, "source_start_sample": 0, "source_end_sample": 207360, "first_word_output_sample": 0, "last_word_output_sample": 203520, "samples": 210720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-timed-u06-c00.mp3" }, "de-conversation-03-base17-segment-timed-u02-c00": { "id": "de-conversation-03-base17-segment-timed-u02-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85133, "plain": "Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf.", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 11, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.7113611282520838 }, "de-conversation-03-base17-segment-free-u04-c00": { "id": "de-conversation-03-base17-segment-free-u04-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85165, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz.", "source_duration_s": 8.88, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 291840, "script_start_s": 0.08, "script_end_s": 6.08, "all_asr_last_word_s": 8.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07692307692307693, "source_identity_reward": 0.6778165193019158, "source_start_sample": 0, "source_end_sample": 299040, "first_word_output_sample": 3840, "last_word_output_sample": 291840, "samples": 299040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.65, "added_silence_s": 0.0, "duration_s": 6.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-free-u04-c00.mp3" }, "en-conversation-03-base17-segment-free-u03-c00": { "id": "en-conversation-03-base17-segment-free-u03-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75149, "plain": "I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 20.4, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 230400, "script_start_s": 0.0, "script_end_s": 4.8, "all_asr_last_word_s": 16.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07142857142857142, "source_identity_reward": 0.10270974365558815, "source_start_sample": 0, "source_end_sample": 237600, "first_word_output_sample": 0, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 15.45, "added_silence_s": 0.0, "duration_s": 4.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-free-u03-c00.mp3" }, "de-conversation-03-base17-segment-timed-u06-c00": { "id": "de-conversation-03-base17-segment-timed-u06-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85197, "plain": "Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie.", "source_duration_s": 4.48, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 9, "coverage": 0.8181818181818182, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.18181818181818182, "source_identity_reward": 0.514278318667513 }, "de-conversation-05-base17-segment-timed-u14-c00": { "id": "de-conversation-05-base17-segment-timed-u14-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87325, "plain": "Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein.", "source_duration_s": 5.68, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 15, "coverage": 0.9375, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 268800, "script_start_s": 0.0, "script_end_s": 5.6, "all_asr_last_word_s": 5.6000000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u14-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1875, "source_identity_reward": 0.5731128083381829, "source_start_sample": 0, "source_end_sample": 272640, "first_word_output_sample": 0, "last_word_output_sample": 268800, "samples": 276000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-timed-u14-c00.mp3" }, "de-conversation-02-base17-segment-free-u05-c00": { "id": "de-conversation-02-base17-segment-free-u05-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84181, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 349440, "script_start_s": 0.08, "script_end_s": 7.28, "all_asr_last_word_s": 7.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5226029451697057, "source_start_sample": 0, "source_end_sample": 356640, "first_word_output_sample": 3840, "last_word_output_sample": 349440, "samples": 356640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 22.57, "added_silence_s": 0.0, "duration_s": 7.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-free-u05-c00.mp3" }, "en-conversation-03-base17-block-timed-u00-c00": { "id": "en-conversation-03-base17-block-timed-u00-c00", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 75601, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 17.12, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 51, "coverage": 0.9807692307692307, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 817920, "script_start_s": 0.0, "script_end_s": 17.04, "all_asr_last_word_s": 17.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.019230769230769232, "source_identity_reward": 0.13188311973964797, "source_start_sample": 0, "source_end_sample": 821760, "first_word_output_sample": 0, "last_word_output_sample": 817920, "samples": 825120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u00-c00.mp3" }, "en-conversation-03-base17-block-timed-u00-c01": { "id": "en-conversation-03-base17-block-timed-u00-c01", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 75602, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 17.04, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 51, "coverage": 0.9807692307692307, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 814080, "script_start_s": 0.24, "script_end_s": 16.96, "all_asr_last_word_s": 16.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.019230769230769232, "source_identity_reward": 0.16386836595722812, "source_start_sample": 7680, "source_end_sample": 817920, "first_word_output_sample": 3840, "last_word_output_sample": 806400, "samples": 813600, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u00-c01.mp3" }, "en-conversation-03-base17-block-timed-u00-c02": { "id": "en-conversation-03-base17-block-timed-u00-c02", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 75603, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 17.04, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 51, "coverage": 0.9807692307692307, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 814080, "script_start_s": 0.16, "script_end_s": 16.96, "all_asr_last_word_s": 16.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.019230769230769232, "source_identity_reward": 0.21025557057848207, "source_start_sample": 3840, "source_end_sample": 817920, "first_word_output_sample": 3840, "last_word_output_sample": 810240, "samples": 817440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u00-c02.mp3" }, "en-conversation-03-base17-block-timed-u00-c03": { "id": "en-conversation-03-base17-block-timed-u00-c03", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 75604, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 17.12, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 817920, "script_start_s": 0.16, "script_end_s": 17.04, "all_asr_last_word_s": 17.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.15442146977993934, "source_start_sample": 3840, "source_end_sample": 821760, "first_word_output_sample": 3840, "last_word_output_sample": 814080, "samples": 821280, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u00-c03.mp3" }, "en-conversation-03-base17-block-timed-u00-c04": { "id": "en-conversation-03-base17-block-timed-u00-c04", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 75605, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 17.04, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 810240, "script_start_s": 0.0, "script_end_s": 16.88, "all_asr_last_word_s": 16.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.1617148926336902, "source_start_sample": 0, "source_end_sample": 817440, "first_word_output_sample": 0, "last_word_output_sample": 810240, "samples": 817440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u00-c04.mp3" }, "en-conversation-03-base17-block-timed-u00-c05": { "id": "en-conversation-03-base17-block-timed-u00-c05", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 75606, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 17.12, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 814080, "script_start_s": 0.0, "script_end_s": 16.96, "all_asr_last_word_s": 16.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.16329992055945042, "source_start_sample": 0, "source_end_sample": 821280, "first_word_output_sample": 0, "last_word_output_sample": 814080, "samples": 821280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u00-c05.mp3" }, "en-conversation-03-base17-block-timed-u00-c06": { "id": "en-conversation-03-base17-block-timed-u00-c06", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 75607, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 17.04, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 51, "coverage": 0.9807692307692307, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 814080, "script_start_s": 0.16, "script_end_s": 16.96, "all_asr_last_word_s": 16.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.019230769230769232, "source_identity_reward": 0.21110483041861625, "source_start_sample": 3840, "source_end_sample": 817920, "first_word_output_sample": 3840, "last_word_output_sample": 810240, "samples": 817440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u00-c06.mp3" }, "en-conversation-03-base17-block-timed-u00-c07": { "id": "en-conversation-03-base17-block-timed-u00-c07", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 75608, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 17.04, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 814080, "script_start_s": 0.0, "script_end_s": 16.96, "all_asr_last_word_s": 16.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.14654379605833728, "source_start_sample": 0, "source_end_sample": 817920, "first_word_output_sample": 0, "last_word_output_sample": 814080, "samples": 821280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u00-c07.mp3" }, "de-conversation-05-base17-segment-free-u04-c00": { "id": "de-conversation-05-base17-segment-free-u04-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87165, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war.", "source_duration_s": 16.24, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 341760, "script_start_s": 0.24, "script_end_s": 7.12, "all_asr_last_word_s": 15.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.1428571428571428, "source_identity_reward": 0.3639118954090968, "source_start_sample": 7680, "source_end_sample": 348960, "first_word_output_sample": 3840, "last_word_output_sample": 334080, "samples": 341280, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 8.97, "added_silence_s": 0.0, "duration_s": 7.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u04-c00.mp3" }, "de-conversation-03-base17-segment-timed-u07-c00": { "id": "de-conversation-03-base17-segment-timed-u07-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85213, "plain": "Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 4.0, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 188160, "script_start_s": 0.24, "script_end_s": 3.92, "all_asr_last_word_s": 3.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4041685321436599, "source_start_sample": 7680, "source_end_sample": 192000, "first_word_output_sample": 3840, "last_word_output_sample": 180480, "samples": 187680, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-timed-u07-c00.mp3" }, "en-conversation-01-base17-block-timed-u01-c00": { "id": "en-conversation-01-base17-block-timed-u01-c00", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 73617, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.76, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 844800, "script_start_s": 0.16, "script_end_s": 17.6, "all_asr_last_word_s": 17.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.13849577663396506, "source_start_sample": 3840, "source_end_sample": 852000, "first_word_output_sample": 3840, "last_word_output_sample": 840960, "samples": 848160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u01-c00.mp3" }, "en-conversation-01-base17-block-timed-u01-c01": { "id": "en-conversation-01-base17-block-timed-u01-c01", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 73618, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.76, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 844800, "script_start_s": 0.16, "script_end_s": 17.6, "all_asr_last_word_s": 17.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.21024636231391183, "source_start_sample": 3840, "source_end_sample": 852000, "first_word_output_sample": 3840, "last_word_output_sample": 840960, "samples": 848160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u01-c01.mp3" }, "en-conversation-01-base17-block-timed-u01-c02": { "id": "en-conversation-01-base17-block-timed-u01-c02", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 73619, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.76, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 848640, "script_start_s": 0.16, "script_end_s": 17.68, "all_asr_last_word_s": 17.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.16126171851719087, "source_start_sample": 3840, "source_end_sample": 852480, "first_word_output_sample": 3840, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u01-c02.mp3" }, "en-conversation-01-base17-block-timed-u01-c03": { "id": "en-conversation-01-base17-block-timed-u01-c03", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 73620, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.76, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 844800, "script_start_s": 0.0, "script_end_s": 17.6, "all_asr_last_word_s": 17.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.17038489316019798, "source_start_sample": 0, "source_end_sample": 852000, "first_word_output_sample": 0, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u01-c03.mp3" }, "en-conversation-01-base17-block-timed-u01-c04": { "id": "en-conversation-01-base17-block-timed-u01-c04", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 73621, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.76, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 848640, "script_start_s": 0.16, "script_end_s": 17.68, "all_asr_last_word_s": 17.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.11408909312385618, "source_start_sample": 3840, "source_end_sample": 852480, "first_word_output_sample": 3840, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u01-c04.mp3" }, "en-conversation-01-base17-block-timed-u01-c05": { "id": "en-conversation-01-base17-block-timed-u01-c05", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 73622, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.76, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 848640, "script_start_s": 0.16, "script_end_s": 17.68, "all_asr_last_word_s": 17.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.10467386722247365, "source_start_sample": 3840, "source_end_sample": 852480, "first_word_output_sample": 3840, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u01-c05.mp3" }, "en-conversation-01-base17-block-timed-u01-c06": { "id": "en-conversation-01-base17-block-timed-u01-c06", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 73623, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.76, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 848640, "script_start_s": 0.16, "script_end_s": 17.68, "all_asr_last_word_s": 17.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.15786018598578294, "source_start_sample": 3840, "source_end_sample": 852480, "first_word_output_sample": 3840, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u01-c06.mp3" }, "en-conversation-01-base17-block-timed-u01-c07": { "id": "en-conversation-01-base17-block-timed-u01-c07", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 73624, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.76, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 848640, "script_start_s": 0.16, "script_end_s": 17.68, "all_asr_last_word_s": 17.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.12206115704881527, "source_start_sample": 3840, "source_end_sample": 852480, "first_word_output_sample": 3840, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u01-c07.mp3" }, "en-conversation-01-base17-segment-timed-u00-c00": { "id": "en-conversation-01-base17-segment-timed-u00-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73101, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large.", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 12, "coverage": 0.8571428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 203520, "script_start_s": 0.24, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.14285714285714285, "source_identity_reward": 0.44700805823436685, "source_start_sample": 7680, "source_end_sample": 207360, "first_word_output_sample": 3840, "last_word_output_sample": 195840, "samples": 203040, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-timed-u00-c00.mp3" }, "en-conversation-03-base17-segment-timed-u07-c00": { "id": "en-conversation-03-base17-segment-timed-u07-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75213, "plain": "Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 3.76, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 176640, "script_start_s": 0.0, "script_end_s": 3.68, "all_asr_last_word_s": 3.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3933709469191532, "source_start_sample": 0, "source_end_sample": 180480, "first_word_output_sample": 0, "last_word_output_sample": 176640, "samples": 183840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-timed-u07-c00.mp3" }, "de-conversation-02-base17-block-free-u00-c00": { "id": "de-conversation-02-base17-block-free-u00-c00", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 84601, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 26.16, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 56, "coverage": 0.9333333333333333, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.11666666666666667, "source_identity_reward": 0.3329241859874268 }, "de-conversation-02-base17-block-free-u00-c01": { "id": "de-conversation-02-base17-block-free-u00-c01", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 84602, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 60, "matched_tokens": 57, "coverage": 0.95, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1086720, "script_start_s": 0.16, "script_end_s": 22.64, "all_asr_last_word_s": 55.800000000000004 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.38333333333333336, "source_identity_reward": 0.18895533689348024, "source_start_sample": 3840, "source_end_sample": 1093920, "first_word_output_sample": 3840, "last_word_output_sample": 1082880, "samples": 1090080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 37.21, "added_silence_s": 0.0, "duration_s": 22.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u00-c01.mp3" }, "de-conversation-02-base17-block-free-u00-c02": { "id": "de-conversation-02-base17-block-free-u00-c02", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 84603, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 24.4, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 55, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1009920, "script_start_s": 0.0, "script_end_s": 21.04, "all_asr_last_word_s": 23.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.21666666666666667, "source_identity_reward": 0.292817827947313, "source_start_sample": 0, "source_end_sample": 1017120, "first_word_output_sample": 0, "last_word_output_sample": 1009920, "samples": 1017120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 3.21, "added_silence_s": 0.0, "duration_s": 21.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u00-c02.mp3" }, "de-conversation-02-base17-block-free-u00-c03": { "id": "de-conversation-02-base17-block-free-u00-c03", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 84604, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 29.28, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 58, "coverage": 0.9666666666666667, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1071360, "script_start_s": 0.16, "script_end_s": 22.32, "all_asr_last_word_s": 29.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.35, "source_identity_reward": 0.2477332734347729, "source_start_sample": 3840, "source_end_sample": 1078560, "first_word_output_sample": 3840, "last_word_output_sample": 1067520, "samples": 1074720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 6.81, "added_silence_s": 0.0, "duration_s": 22.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u00-c03.mp3" }, "de-conversation-02-base17-block-free-u00-c04": { "id": "de-conversation-02-base17-block-free-u00-c04", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 84605, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 60, "matched_tokens": 56, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1094400, "script_start_s": 0.16, "script_end_s": 22.8, "all_asr_last_word_s": 39.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.13333333333333333, "source_identity_reward": 0.20715718163790392, "source_start_sample": 3840, "source_end_sample": 1101600, "first_word_output_sample": 3840, "last_word_output_sample": 1090560, "samples": 1097760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 37.05, "added_silence_s": 0.0, "duration_s": 22.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u00-c04.mp3" }, "de-conversation-02-base17-block-free-u00-c05": { "id": "de-conversation-02-base17-block-free-u00-c05", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 84606, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 60, "matched_tokens": 55, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1075200, "script_start_s": 0.16, "script_end_s": 22.4, "all_asr_last_word_s": 27.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.35, "source_identity_reward": 0.21133952949372664, "source_start_sample": 3840, "source_end_sample": 1082400, "first_word_output_sample": 3840, "last_word_output_sample": 1071360, "samples": 1078560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 37.45, "added_silence_s": 0.0, "duration_s": 22.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u00-c05.mp3" }, "de-conversation-02-base17-block-free-u00-c06": { "id": "de-conversation-02-base17-block-free-u00-c06", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 84607, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 60, "matched_tokens": 57, "coverage": 0.95, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1094400, "script_start_s": 0.0, "script_end_s": 22.8, "all_asr_last_word_s": 59.879999999999995 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3333333333333333, "source_identity_reward": 0.16280252769627335, "source_start_sample": 0, "source_end_sample": 1101600, "first_word_output_sample": 0, "last_word_output_sample": 1094400, "samples": 1101600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 37.05, "added_silence_s": 0.0, "duration_s": 22.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u00-c06.mp3" }, "de-conversation-02-base17-block-free-u00-c07": { "id": "de-conversation-02-base17-block-free-u00-c07", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 84608, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 28.72, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 57, "coverage": 0.95, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1367040, "script_start_s": 0.0, "script_end_s": 28.48, "all_asr_last_word_s": 28.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3, "source_identity_reward": 0.28544790863216823, "source_start_sample": 0, "source_end_sample": 1374240, "first_word_output_sample": 0, "last_word_output_sample": 1367040, "samples": 1374240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 28.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u00-c07.mp3" }, "en-conversation-05-base17-segment-timed-u09-c00": { "id": "en-conversation-05-base17-segment-timed-u09-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 192000, "script_start_s": 0.16, "script_end_s": 4.0, "all_asr_last_word_s": 4.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.13306359282165137, "source_start_sample": 3840, "source_end_sample": 195840, "first_word_output_sample": 3840, "last_word_output_sample": 188160, "samples": 195360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-timed-u09-c00.mp3" }, "en-conversation-01-base17-block-free-u00-c00": { "id": "en-conversation-01-base17-block-free-u00-c00", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 73601, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 837120, "script_start_s": 0.16, "script_end_s": 17.44, "all_asr_last_word_s": 17.439999999999998 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.19000517220586152, "source_start_sample": 3840, "source_end_sample": 844320, "first_word_output_sample": 3840, "last_word_output_sample": 833280, "samples": 840480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 42.41, "added_silence_s": 0.0, "duration_s": 17.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u00-c00.mp3" }, "en-conversation-01-base17-block-free-u00-c01": { "id": "en-conversation-01-base17-block-free-u00-c01", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 73602, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 825600, "script_start_s": 0.0, "script_end_s": 17.2, "all_asr_last_word_s": 17.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.18551399524680862, "source_start_sample": 0, "source_end_sample": 832800, "first_word_output_sample": 0, "last_word_output_sample": 825600, "samples": 832800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 42.65, "added_silence_s": 0.0, "duration_s": 17.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u00-c01.mp3" }, "en-conversation-01-base17-block-free-u00-c02": { "id": "en-conversation-01-base17-block-free-u00-c02", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 73603, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 831360, "script_start_s": 0.0, "script_end_s": 17.32, "all_asr_last_word_s": 17.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.14624941072030997, "source_start_sample": 0, "source_end_sample": 838560, "first_word_output_sample": 0, "last_word_output_sample": 831360, "samples": 838560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 42.53, "added_silence_s": 0.0, "duration_s": 17.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u00-c02.mp3" }, "en-conversation-01-base17-block-free-u00-c03": { "id": "en-conversation-01-base17-block-free-u00-c03", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 73604, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 52, "matched_tokens": 28, "coverage": 0.5384615384615384, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.46153846153846156, "source_identity_reward": 0.051796484512720155 }, "en-conversation-01-base17-block-free-u00-c04": { "id": "en-conversation-01-base17-block-free-u00-c04", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 73605, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 52, "matched_tokens": 51, "coverage": 0.9807692307692307, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 864000, "script_start_s": 0.0, "script_end_s": 18.0, "all_asr_last_word_s": 39.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.28846153846153844, "source_identity_reward": 0.13545786523397452, "source_start_sample": 0, "source_end_sample": 871200, "first_word_output_sample": 0, "last_word_output_sample": 864000, "samples": 871200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 41.85, "added_silence_s": 0.0, "duration_s": 18.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u00-c04.mp3" }, "en-conversation-01-base17-block-free-u00-c05": { "id": "en-conversation-01-base17-block-free-u00-c05", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 73606, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 19.84, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 860160, "script_start_s": 0.0, "script_end_s": 17.92, "all_asr_last_word_s": 17.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07692307692307693, "source_identity_reward": 0.13616396084621005, "source_start_sample": 0, "source_end_sample": 867360, "first_word_output_sample": 0, "last_word_output_sample": 860160, "samples": 867360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.77, "added_silence_s": 0.0, "duration_s": 18.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u00-c05.mp3" }, "en-conversation-01-base17-block-free-u00-c06": { "id": "en-conversation-01-base17-block-free-u00-c06", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 73607, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 15.6, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 21, "coverage": 0.40384615384615385, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5961538461538461, "source_identity_reward": 0.04857216939027001 }, "en-conversation-01-base17-block-free-u00-c07": { "id": "en-conversation-01-base17-block-free-u00-c07", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 73608, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 17.68, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 49, "coverage": 0.9423076923076923, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 837120, "script_start_s": 0.16, "script_end_s": 17.44, "all_asr_last_word_s": 17.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.057692307692307696, "source_identity_reward": 0.19018641560382865, "source_start_sample": 3840, "source_end_sample": 844320, "first_word_output_sample": 3840, "last_word_output_sample": 833280, "samples": 840480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 17.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u00-c07.mp3" }, "de-conversation-05-base17-segment-free-u03-c00": { "id": "de-conversation-05-base17-segment-free-u03-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87149, "plain": "Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 14.16, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 407040, "script_start_s": 0.16, "script_end_s": 8.48, "all_asr_last_word_s": 10.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.23076923076923078, "source_identity_reward": 0.4395905743781808, "source_start_sample": 3840, "source_end_sample": 414240, "first_word_output_sample": 3840, "last_word_output_sample": 403200, "samples": 410400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 5.53, "added_silence_s": 0.0, "duration_s": 8.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u03-c00.mp3" }, "de-conversation-05-base17-segment-free-u11-c00": { "id": "de-conversation-05-base17-segment-free-u11-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 314880, "script_start_s": 0.24, "script_end_s": 6.56, "all_asr_last_word_s": 21.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.9230769230769231, "source_identity_reward": 0.26865347735631784, "source_start_sample": 7680, "source_end_sample": 322080, "first_word_output_sample": 3840, "last_word_output_sample": 307200, "samples": 314400, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 23.29, "added_silence_s": 0.0, "duration_s": 6.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u11-c00.mp3" }, "de-conversation-05-base17-segment-free-u02-c00": { "id": "de-conversation-05-base17-segment-free-u02-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87133, "plain": "Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen.", "source_duration_s": 7.68, "source_frame_cap": false, "alignment": { "expected_tokens": 10, "matched_tokens": 10, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7076839676530277, "source_start_sample": 0, "source_end_sample": 226080, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.97, "added_silence_s": 0.0, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u02-c00.mp3" }, "de-conversation-03-base17-segment-timed-u04-c00": { "id": "de-conversation-03-base17-segment-timed-u04-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85165, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7687184285862123, "source_start_sample": 0, "source_end_sample": 222720, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-timed-u04-c00.mp3" }, "en-conversation-05-base17-segment-timed-u06-c00": { "id": "en-conversation-05-base17-segment-timed-u06-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77197, "plain": "Exactly. We publish the full scripts and seeds so the exciting failures remain visible too.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 13, "coverage": 0.8666666666666667, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.13333333333333333, "source_identity_reward": 0.3604092506674956 }, "en-conversation-02-base17-segment-free-u10-c00": { "id": "en-conversation-02-base17-segment-free-u10-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 249600, "script_start_s": 0.0, "script_end_s": 5.2, "all_asr_last_word_s": 24.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.5, "source_identity_reward": 0.16821906392297672, "source_start_sample": 0, "source_end_sample": 256800, "first_word_output_sample": 0, "last_word_output_sample": 249600, "samples": 256800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.65, "added_silence_s": 0.0, "duration_s": 5.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-free-u10-c00.mp3" }, "de-conversation-02-base17-segment-free-u09-c00": { "id": "de-conversation-02-base17-segment-free-u09-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 27.36, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 299520, "script_start_s": 0.0, "script_end_s": 6.24, "all_asr_last_word_s": 20.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.35714285714285715, "source_identity_reward": 0.46737825674466976, "source_start_sample": 0, "source_end_sample": 306720, "first_word_output_sample": 0, "last_word_output_sample": 299520, "samples": 306720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 20.97, "added_silence_s": 0.0, "duration_s": 6.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-free-u09-c00.mp3" }, "de-conversation-02-base17-segment-timed-u02-c00": { "id": "de-conversation-02-base17-segment-timed-u02-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84133, "plain": "Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog.", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 8, "coverage": 0.6666666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 203520, "script_start_s": 0.0, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4166666666666667, "source_identity_reward": 0.47406635000448555, "source_start_sample": 0, "source_end_sample": 207360, "first_word_output_sample": 0, "last_word_output_sample": 203520, "samples": 210720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-timed-u02-c00.mp3" }, "en-conversation-03-base17-segment-timed-u11-c00": { "id": "en-conversation-03-base17-segment-timed-u11-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 230400, "script_start_s": 0.16, "script_end_s": 4.8, "all_asr_last_word_s": 4.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.29443945819864514, "source_start_sample": 3840, "source_end_sample": 237600, "first_word_output_sample": 3840, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-timed-u11-c00.mp3" }, "de-conversation-05-base17-segment-free-u01-c00": { "id": "de-conversation-05-base17-segment-free-u01-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87117, "plain": "Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus?", "source_duration_s": 10.08, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 410880, "script_start_s": 0.24, "script_end_s": 8.56, "all_asr_last_word_s": 8.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.49130606914672814, "source_start_sample": 7680, "source_end_sample": 418080, "first_word_output_sample": 3840, "last_word_output_sample": 403200, "samples": 410400, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 1.37, "added_silence_s": 0.0, "duration_s": 8.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u01-c00.mp3" }, "en-conversation-03-base17-segment-free-u05-c00": { "id": "en-conversation-03-base17-segment-free-u05-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75181, "plain": "Good. I want the delighted host to remain the same person when she becomes doubtful.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 299520, "script_start_s": 0.16, "script_end_s": 6.24, "all_asr_last_word_s": 11.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06666666666666667, "source_identity_reward": 0.27938452043338263, "source_start_sample": 3840, "source_end_sample": 306720, "first_word_output_sample": 3840, "last_word_output_sample": 295680, "samples": 302880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.61, "added_silence_s": 0.0, "duration_s": 6.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-free-u05-c00.mp3" }, "de-conversation-03-base17-segment-free-u08-c00": { "id": "de-conversation-03-base17-segment-free-u08-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 334080, "script_start_s": 0.0, "script_end_s": 6.96, "all_asr_last_word_s": 6.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7209173155326418, "source_start_sample": 0, "source_end_sample": 341280, "first_word_output_sample": 0, "last_word_output_sample": 334080, "samples": 341280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 22.89, "added_silence_s": 0.0, "duration_s": 7.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-free-u08-c00.mp3" }, "en-conversation-02-base17-segment-free-u03-c00": { "id": "en-conversation-02-base17-segment-free-u03-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74149, "plain": "And does the second host interrupt with healthy skepticism, or just politely agree?", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 284160, "script_start_s": 0.16, "script_end_s": 5.92, "all_asr_last_word_s": 24.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0, "source_identity_reward": 0.19132956408500676, "source_start_sample": 3840, "source_end_sample": 291360, "first_word_output_sample": 3840, "last_word_output_sample": 280320, "samples": 287520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.93, "added_silence_s": 0.0, "duration_s": 5.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-free-u03-c00.mp3" }, "en-conversation-03-base17-segment-free-u07-c00": { "id": "en-conversation-03-base17-segment-free-u07-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75213, "plain": "Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 8.0, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 257280, "script_start_s": 0.0, "script_end_s": 5.36, "all_asr_last_word_s": 5.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3531178672145877, "source_start_sample": 0, "source_end_sample": 264480, "first_word_output_sample": 0, "last_word_output_sample": 257280, "samples": 264480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.49, "added_silence_s": 0.0, "duration_s": 5.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-free-u07-c00.mp3" }, "de-conversation-05-base17-segment-timed-u09-c00": { "id": "de-conversation-05-base17-segment-timed-u09-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 12, "coverage": 0.8571428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 230400, "script_start_s": 0.0, "script_end_s": 4.8, "all_asr_last_word_s": 4.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.14285714285714285, "source_identity_reward": 0.5117153965509218, "source_start_sample": 0, "source_end_sample": 237600, "first_word_output_sample": 0, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-timed-u09-c00.mp3" }, "en-conversation-05-base17-segment-timed-u10-c00": { "id": "en-conversation-05-base17-segment-timed-u10-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 218880, "script_start_s": 0.16, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3409356549859705, "source_start_sample": 3840, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-timed-u10-c00.mp3" }, "en-conversation-02-base17-segment-timed-u00-c00": { "id": "en-conversation-02-base17-segment-timed-u00-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74101, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 218880, "script_start_s": 0.16, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.43540424143025436, "source_start_sample": 3840, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-timed-u00-c00.mp3" }, "en-conversation-01-base17-block-free-u01-c00": { "id": "en-conversation-01-base17-block-free-u01-c00", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 73617, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 25.28, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 925440, "script_start_s": 0.0, "script_end_s": 19.28, "all_asr_last_word_s": 24.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3018867924528302, "source_identity_reward": 0.1849783944920829, "source_start_sample": 0, "source_end_sample": 932640, "first_word_output_sample": 0, "last_word_output_sample": 925440, "samples": 932640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 5.85, "added_silence_s": 0.0, "duration_s": 19.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u01-c00.mp3" }, "en-conversation-01-base17-block-free-u01-c01": { "id": "en-conversation-01-base17-block-free-u01-c01", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 73618, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 21.92, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 994560, "script_start_s": 0.16, "script_end_s": 20.72, "all_asr_last_word_s": 20.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.20480687472299514, "source_start_sample": 3840, "source_end_sample": 1001760, "first_word_output_sample": 3840, "last_word_output_sample": 990720, "samples": 997920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.05, "added_silence_s": 0.0, "duration_s": 20.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u01-c01.mp3" }, "en-conversation-01-base17-block-free-u01-c02": { "id": "en-conversation-01-base17-block-free-u01-c02", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 73619, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 53, "matched_tokens": 51, "coverage": 0.9622641509433962, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 902400, "script_start_s": 0.24, "script_end_s": 18.8, "all_asr_last_word_s": 37.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.22641509433962265, "source_identity_reward": 0.18375862710942506, "source_start_sample": 7680, "source_end_sample": 909600, "first_word_output_sample": 3840, "last_word_output_sample": 894720, "samples": 901920, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 41.05, "added_silence_s": 0.0, "duration_s": 18.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u01-c02.mp3" }, "en-conversation-01-base17-block-free-u01-c03": { "id": "en-conversation-01-base17-block-free-u01-c03", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 73620, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 983040, "script_start_s": 0.16, "script_end_s": 20.48, "all_asr_last_word_s": 20.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.11049364644839489, "source_start_sample": 3840, "source_end_sample": 990240, "first_word_output_sample": 3840, "last_word_output_sample": 979200, "samples": 986400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 39.37, "added_silence_s": 0.0, "duration_s": 20.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u01-c03.mp3" }, "en-conversation-01-base17-block-free-u01-c04": { "id": "en-conversation-01-base17-block-free-u01-c04", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 73621, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 20.88, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 956160, "script_start_s": 0.16, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.19414622202603218, "source_start_sample": 3840, "source_end_sample": 963360, "first_word_output_sample": 3840, "last_word_output_sample": 952320, "samples": 959520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.81, "added_silence_s": 0.0, "duration_s": 19.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u01-c04.mp3" }, "en-conversation-01-base17-block-free-u01-c05": { "id": "en-conversation-01-base17-block-free-u01-c05", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 73622, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 879360, "script_start_s": 0.24, "script_end_s": 18.32, "all_asr_last_word_s": 23.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3018867924528302, "source_identity_reward": 0.0867709317327938, "source_start_sample": 7680, "source_end_sample": 886560, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 41.53, "added_silence_s": 0.0, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u01-c05.mp3" }, "en-conversation-01-base17-block-free-u01-c06": { "id": "en-conversation-01-base17-block-free-u01-c06", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 73623, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 902400, "script_start_s": 0.16, "script_end_s": 18.8, "all_asr_last_word_s": 18.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.20622246927419496, "source_start_sample": 3840, "source_end_sample": 909600, "first_word_output_sample": 3840, "last_word_output_sample": 898560, "samples": 905760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 41.05, "added_silence_s": 0.0, "duration_s": 18.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-free-u01-c06.mp3" }, "en-conversation-01-base17-block-free-u01-c07": { "id": "en-conversation-01-base17-block-free-u01-c07", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 73624, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 53, "matched_tokens": 33, "coverage": 0.6226415094339622, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.37735849056603776, "source_identity_reward": 0.07359942510215446 }, "de-conversation-05-base17-segment-timed-u04-c00": { "id": "de-conversation-05-base17-segment-timed-u04-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87165, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 234240, "script_start_s": 0.24, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7532410280986797, "source_start_sample": 7680, "source_end_sample": 238080, "first_word_output_sample": 3840, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-timed-u04-c00.mp3" }, "de-conversation-02-base17-segment-timed-u11-c00": { "id": "de-conversation-02-base17-segment-timed-u11-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 218880, "script_start_s": 0.16, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.670745519840223, "source_start_sample": 3840, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-timed-u11-c00.mp3" }, "en-conversation-05-base17-segment-free-u06-c00": { "id": "en-conversation-05-base17-segment-free-u06-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77197, "plain": "Exactly. We publish the full scripts and seeds so the exciting failures remain visible too.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 322560, "script_start_s": 0.24, "script_end_s": 6.72, "all_asr_last_word_s": 29.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.8666666666666667, "source_identity_reward": 0.2440713689255135, "source_start_sample": 7680, "source_end_sample": 329760, "first_word_output_sample": 3840, "last_word_output_sample": 314880, "samples": 322080, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 23.13, "added_silence_s": 0.0, "duration_s": 6.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-free-u06-c00.mp3" }, "en-conversation-05-base17-segment-timed-u05-c00": { "id": "en-conversation-05-base17-segment-timed-u05-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77181, "plain": "I want surprise, warmth, a little teasing, and no one reading the stage directions aloud.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 234240, "script_start_s": 0.0, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.10650760705934556, "source_start_sample": 0, "source_end_sample": 238080, "first_word_output_sample": 0, "last_word_output_sample": 234240, "samples": 241440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-timed-u05-c00.mp3" }, "en-conversation-05-base17-segment-timed-u08-c00": { "id": "en-conversation-05-base17-segment-timed-u08-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 218880, "script_start_s": 0.16, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.21358509179512433, "source_start_sample": 3840, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-timed-u08-c00.mp3" }, "en-conversation-01-base17-block-timed-u02-c00": { "id": "en-conversation-01-base17-block-timed-u02-c00", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 73633, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.09401887329379818, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u02-c00.mp3" }, "en-conversation-01-base17-block-timed-u02-c01": { "id": "en-conversation-01-base17-block-timed-u02-c01", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 73634, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1558084548614977, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u02-c01.mp3" }, "en-conversation-01-base17-block-timed-u02-c02": { "id": "en-conversation-01-base17-block-timed-u02-c02", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 73635, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 875520, "script_start_s": 0.16, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.11173033361849942, "source_start_sample": 3840, "source_end_sample": 879360, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u02-c02.mp3" }, "en-conversation-01-base17-block-timed-u02-c03": { "id": "en-conversation-01-base17-block-timed-u02-c03", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 73636, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 879360, "script_start_s": 0.24, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.11163125893617795, "source_start_sample": 7680, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u02-c03.mp3" }, "en-conversation-01-base17-block-timed-u02-c04": { "id": "en-conversation-01-base17-block-timed-u02-c04", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 73637, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1710555683214304, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u02-c04.mp3" }, "en-conversation-01-base17-block-timed-u02-c05": { "id": "en-conversation-01-base17-block-timed-u02-c05", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 73638, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.09967764320542052, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u02-c05.mp3" }, "en-conversation-01-base17-block-timed-u02-c06": { "id": "en-conversation-01-base17-block-timed-u02-c06", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 73639, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.15405932398020714, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u02-c06.mp3" }, "en-conversation-01-base17-block-timed-u02-c07": { "id": "en-conversation-01-base17-block-timed-u02-c07", "case": "en-conversation-01", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 73640, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.10447043938315546, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-block-timed-u02-c07.mp3" }, "de-conversation-05-base17-segment-free-u07-c00": { "id": "de-conversation-05-base17-segment-free-u07-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87213, "plain": "Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 261120, "script_start_s": 0.0, "script_end_s": 5.44, "all_asr_last_word_s": 5.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5181206324178877, "source_start_sample": 0, "source_end_sample": 268320, "first_word_output_sample": 0, "last_word_output_sample": 261120, "samples": 268320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.41, "added_silence_s": 0.0, "duration_s": 5.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u07-c00.mp3" }, "de-conversation-04-base17-segment-timed-u04-c00": { "id": "de-conversation-04-base17-segment-timed-u04-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86165, "plain": "Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 222720, "script_start_s": 0.0, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7322764782560408, "source_start_sample": 0, "source_end_sample": 226560, "first_word_output_sample": 0, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-timed-u04-c00.mp3" }, "en-conversation-05-base17-segment-free-u08-c00": { "id": "en-conversation-05-base17-segment-free-u08-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 272640, "script_start_s": 0.16, "script_end_s": 5.68, "all_asr_last_word_s": 26.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0, "source_identity_reward": 0.17244944266529794, "source_start_sample": 3840, "source_end_sample": 279840, "first_word_output_sample": 3840, "last_word_output_sample": 268800, "samples": 276000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 24.17, "added_silence_s": 0.0, "duration_s": 5.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-free-u08-c00.mp3" }, "de-conversation-01-base17-segment-timed-u04-c00": { "id": "de-conversation-01-base17-segment-timed-u04-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83165, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt.", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 11, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 207360, "script_start_s": 0.08, "script_end_s": 4.32, "all_asr_last_word_s": 4.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.6917718659992432, "source_start_sample": 0, "source_end_sample": 211200, "first_word_output_sample": 3840, "last_word_output_sample": 207360, "samples": 214560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-timed-u04-c00.mp3" }, "en-conversation-04-base17-segment-free-u08-c00": { "id": "en-conversation-04-base17-segment-free-u08-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 10.56, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 13, "coverage": 0.9285714285714286, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 303360, "script_start_s": 0.0, "script_end_s": 6.32, "all_asr_last_word_s": 6.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07142857142857142, "source_identity_reward": 0.2980248477145045, "source_start_sample": 0, "source_end_sample": 310560, "first_word_output_sample": 0, "last_word_output_sample": 303360, "samples": 310560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.09, "added_silence_s": 0.0, "duration_s": 6.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-free-u08-c00.mp3" }, "en-conversation-02-base17-segment-timed-u10-c00": { "id": "en-conversation-02-base17-segment-timed-u10-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 218880, "script_start_s": 0.16, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.38576696519919096, "source_start_sample": 3840, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-timed-u10-c00.mp3" }, "de-conversation-03-base17-segment-free-u01-c00": { "id": "de-conversation-03-base17-segment-free-u01-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85117, "plain": "Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt?", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 307200, "script_start_s": 0.0, "script_end_s": 6.4, "all_asr_last_word_s": 29.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.8461538461538461, "source_identity_reward": 0.39010800983031535, "source_start_sample": 0, "source_end_sample": 314400, "first_word_output_sample": 0, "last_word_output_sample": 307200, "samples": 314400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.45, "added_silence_s": 0.0, "duration_s": 6.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-free-u01-c00.mp3" }, "de-conversation-02-base17-segment-timed-u07-c00": { "id": "de-conversation-02-base17-segment-timed-u07-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84213, "plain": "Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 215040, "script_start_s": 0.0, "script_end_s": 4.48, "all_asr_last_word_s": 4.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5476021843931138, "source_start_sample": 0, "source_end_sample": 222240, "first_word_output_sample": 0, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-timed-u07-c00.mp3" }, "en-conversation-04-base17-block-timed-u00-c00": { "id": "en-conversation-04-base17-block-timed-u00-c00", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 76601, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.4, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 66, "coverage": 0.9705882352941176, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1067520, "script_start_s": 0.16, "script_end_s": 22.24, "all_asr_last_word_s": 22.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029411764705882353, "source_identity_reward": 0.21446653896974333, "source_start_sample": 3840, "source_end_sample": 1074720, "first_word_output_sample": 3840, "last_word_output_sample": 1063680, "samples": 1070880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u00-c00.mp3" }, "en-conversation-04-base17-block-timed-u00-c01": { "id": "en-conversation-04-base17-block-timed-u00-c01", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 76602, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.4, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1067520, "script_start_s": 0.16, "script_end_s": 22.24, "all_asr_last_word_s": 22.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.17993356456020845, "source_start_sample": 3840, "source_end_sample": 1074720, "first_word_output_sample": 3840, "last_word_output_sample": 1063680, "samples": 1070880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u00-c01.mp3" }, "en-conversation-04-base17-block-timed-u00-c02": { "id": "en-conversation-04-base17-block-timed-u00-c02", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 76603, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.4, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1071360, "script_start_s": 0.16, "script_end_s": 22.32, "all_asr_last_word_s": 22.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.21944892279965555, "source_start_sample": 3840, "source_end_sample": 1075200, "first_word_output_sample": 3840, "last_word_output_sample": 1067520, "samples": 1074720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 22.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u00-c02.mp3" }, "en-conversation-04-base17-block-timed-u00-c03": { "id": "en-conversation-04-base17-block-timed-u00-c03", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 76604, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.4, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1067520, "script_start_s": 0.16, "script_end_s": 22.24, "all_asr_last_word_s": 22.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.11545576090975662, "source_start_sample": 3840, "source_end_sample": 1074720, "first_word_output_sample": 3840, "last_word_output_sample": 1063680, "samples": 1070880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u00-c03.mp3" }, "en-conversation-04-base17-block-timed-u00-c04": { "id": "en-conversation-04-base17-block-timed-u00-c04", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 76605, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.4, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 64, "coverage": 0.9411764705882353, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1071360, "script_start_s": 0.16, "script_end_s": 22.32, "all_asr_last_word_s": 22.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.058823529411764705, "source_identity_reward": 0.2366683057803668, "source_start_sample": 3840, "source_end_sample": 1075200, "first_word_output_sample": 3840, "last_word_output_sample": 1067520, "samples": 1074720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 22.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u00-c04.mp3" }, "en-conversation-04-base17-block-timed-u00-c05": { "id": "en-conversation-04-base17-block-timed-u00-c05", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 76606, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.4, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1071360, "script_start_s": 0.16, "script_end_s": 22.32, "all_asr_last_word_s": 22.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.19241987924343945, "source_start_sample": 3840, "source_end_sample": 1075200, "first_word_output_sample": 3840, "last_word_output_sample": 1067520, "samples": 1074720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 22.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u00-c05.mp3" }, "en-conversation-04-base17-block-timed-u00-c06": { "id": "en-conversation-04-base17-block-timed-u00-c06", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 76607, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.4, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1071360, "script_start_s": 0.0, "script_end_s": 22.32, "all_asr_last_word_s": 22.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.16460880490432014, "source_start_sample": 0, "source_end_sample": 1075200, "first_word_output_sample": 0, "last_word_output_sample": 1071360, "samples": 1078560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 22.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u00-c06.mp3" }, "en-conversation-04-base17-block-timed-u00-c07": { "id": "en-conversation-04-base17-block-timed-u00-c07", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 76608, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.4, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1071360, "script_start_s": 0.16, "script_end_s": 22.32, "all_asr_last_word_s": 22.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.22228559372952192, "source_start_sample": 3840, "source_end_sample": 1075200, "first_word_output_sample": 3840, "last_word_output_sample": 1067520, "samples": 1074720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 22.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u00-c07.mp3" }, "de-conversation-03-base17-segment-timed-u00-c00": { "id": "de-conversation-03-base17-segment-timed-u00-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85101, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können.", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 9, "coverage": 0.75, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3333333333333333, "source_identity_reward": 0.38631065246032104 }, "de-conversation-03-base17-segment-free-u06-c00": { "id": "de-conversation-03-base17-segment-free-u06-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85197, "plain": "Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 387840, "script_start_s": 0.16, "script_end_s": 8.08, "all_asr_last_word_s": 8.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7271304730473348, "source_start_sample": 3840, "source_end_sample": 395040, "first_word_output_sample": 3840, "last_word_output_sample": 384000, "samples": 391200, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 21.77, "added_silence_s": 0.0, "duration_s": 8.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-free-u06-c00.mp3" }, "de-conversation-02-base17-segment-free-u03-c00": { "id": "de-conversation-02-base17-segment-free-u03-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84149, "plain": "Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen?", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 318720, "script_start_s": 0.08, "script_end_s": 6.64, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.6831869307503319, "source_start_sample": 0, "source_end_sample": 325920, "first_word_output_sample": 3840, "last_word_output_sample": 318720, "samples": 325920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.21, "added_silence_s": 0.0, "duration_s": 6.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-free-u03-c00.mp3" }, "de-conversation-01-base17-segment-timed-u10-c00": { "id": "de-conversation-01-base17-segment-timed-u10-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 5.68, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 264960, "script_start_s": 0.16, "script_end_s": 5.52, "all_asr_last_word_s": 5.5200000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6954813167155617, "source_start_sample": 3840, "source_end_sample": 272160, "first_word_output_sample": 3840, "last_word_output_sample": 261120, "samples": 268320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-timed-u10-c00.mp3" }, "en-conversation-05-base17-segment-timed-u14-c00": { "id": "en-conversation-05-base17-segment-timed-u14-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77325, "plain": "Exactly. The exchange should feel responsive, with the same two people returning after every change of mood.", "source_duration_s": 5.6, "source_frame_cap": false, "alignment": { "expected_tokens": 17, "matched_tokens": 17, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 264960, "script_start_s": 0.24, "script_end_s": 5.52, "all_asr_last_word_s": 5.5200000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u14-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4718032396326638, "source_start_sample": 7680, "source_end_sample": 268800, "first_word_output_sample": 3840, "last_word_output_sample": 257280, "samples": 264480, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-timed-u14-c00.mp3" }, "en-conversation-01-base17-segment-free-u05-c00": { "id": "en-conversation-01-base17-segment-free-u05-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73181, "plain": "That distinction makes me trust the demonstration a little more. Annoyingly sensible of you.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 825600, "script_start_s": 0.0, "script_end_s": 17.2, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2857142857142857, "source_identity_reward": 0.17324496720055366, "source_start_sample": 0, "source_end_sample": 832800, "first_word_output_sample": 0, "last_word_output_sample": 825600, "samples": 832800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 12.65, "added_silence_s": 0.0, "duration_s": 17.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-free-u05-c00.mp3" }, "de-conversation-03-base17-segment-free-u05-c00": { "id": "de-conversation-03-base17-segment-free-u05-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85181, "plain": "Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 311040, "script_start_s": 0.16, "script_end_s": 6.48, "all_asr_last_word_s": 6.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.710905971577488, "source_start_sample": 3840, "source_end_sample": 318240, "first_word_output_sample": 3840, "last_word_output_sample": 307200, "samples": 314400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.37, "added_silence_s": 0.0, "duration_s": 6.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-free-u05-c00.mp3" }, "en-conversation-04-base17-segment-timed-u11-c00": { "id": "en-conversation-04-base17-segment-timed-u11-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 234240, "script_start_s": 0.24, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3692129699782127, "source_start_sample": 7680, "source_end_sample": 238080, "first_word_output_sample": 3840, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-timed-u11-c00.mp3" }, "de-conversation-02-base17-segment-timed-u09-c00": { "id": "de-conversation-02-base17-segment-timed-u09-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 230400, "script_start_s": 0.0, "script_end_s": 4.8, "all_asr_last_word_s": 4.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.29430303817036496, "source_start_sample": 0, "source_end_sample": 237600, "first_word_output_sample": 0, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-timed-u09-c00.mp3" }, "en-conversation-03-base17-segment-timed-u08-c00": { "id": "en-conversation-03-base17-segment-timed-u08-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 218880, "script_start_s": 0.16, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.15626849657755965, "source_start_sample": 3840, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-timed-u08-c00.mp3" }, "de-conversation-01-base17-segment-free-u11-c00": { "id": "de-conversation-01-base17-segment-free-u11-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 337920, "script_start_s": 0.24, "script_end_s": 7.04, "all_asr_last_word_s": 15.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.46153846153846156, "source_identity_reward": 0.3464922866546678, "source_start_sample": 7680, "source_end_sample": 345120, "first_word_output_sample": 3840, "last_word_output_sample": 330240, "samples": 337440, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 22.81, "added_silence_s": 0.0, "duration_s": 7.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-free-u11-c00.mp3" }, "en-conversation-05-base17-segment-free-u15-c00": { "id": "en-conversation-05-base17-segment-free-u15-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77341, "plain": "Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 8.96, "source_frame_cap": false, "alignment": { "expected_tokens": 19, "matched_tokens": 19, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 307200, "script_start_s": 0.16, "script_end_s": 6.4, "all_asr_last_word_s": 6.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u15-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.0846590943552632, "source_start_sample": 3840, "source_end_sample": 314400, "first_word_output_sample": 3840, "last_word_output_sample": 303360, "samples": 310560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 2.41, "added_silence_s": 0.0, "duration_s": 6.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-free-u15-c00.mp3" }, "en-conversation-04-base17-segment-timed-u00-c00": { "id": "en-conversation-04-base17-segment-timed-u00-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76101, "plain": "I can write a requested length next to an individual sentence in the script.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.597352668524356, "source_start_sample": 0, "source_end_sample": 222720, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-timed-u00-c00.mp3" }, "de-conversation-03-base17-block-free-u00-c00": { "id": "de-conversation-03-base17-block-free-u00-c00", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 85601, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 913920, "script_start_s": 0.08, "script_end_s": 19.04, "all_asr_last_word_s": 19.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.315188545182131, "source_start_sample": 0, "source_end_sample": 921120, "first_word_output_sample": 3840, "last_word_output_sample": 913920, "samples": 921120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 40.81, "added_silence_s": 0.0, "duration_s": 19.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u00-c00.mp3" }, "de-conversation-03-base17-block-free-u00-c01": { "id": "de-conversation-03-base17-block-free-u00-c01", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 85602, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 18.8, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 48, "coverage": 0.9795918367346939, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 890880, "script_start_s": 0.08, "script_end_s": 18.56, "all_asr_last_word_s": 18.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02040816326530612, "source_identity_reward": 0.20086264412192698, "source_start_sample": 0, "source_end_sample": 898080, "first_word_output_sample": 3840, "last_word_output_sample": 890880, "samples": 898080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 18.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u00-c01.mp3" }, "de-conversation-03-base17-block-free-u00-c02": { "id": "de-conversation-03-base17-block-free-u00-c02", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 85603, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 21.28, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 967680, "script_start_s": 0.0, "script_end_s": 20.16, "all_asr_last_word_s": 20.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.20665923521185112, "source_start_sample": 0, "source_end_sample": 974880, "first_word_output_sample": 0, "last_word_output_sample": 967680, "samples": 974880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.97, "added_silence_s": 0.0, "duration_s": 20.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u00-c02.mp3" }, "de-conversation-03-base17-block-free-u00-c03": { "id": "de-conversation-03-base17-block-free-u00-c03", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 85604, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 19.84, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 929280, "script_start_s": 0.16, "script_end_s": 19.36, "all_asr_last_word_s": 19.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3570814081982742, "source_start_sample": 3840, "source_end_sample": 936480, "first_word_output_sample": 3840, "last_word_output_sample": 925440, "samples": 932640, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.33, "added_silence_s": 0.0, "duration_s": 19.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u00-c03.mp3" }, "de-conversation-03-base17-block-free-u00-c04": { "id": "de-conversation-03-base17-block-free-u00-c04", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 85605, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 931200, "script_start_s": 0.16, "script_end_s": 19.4, "all_asr_last_word_s": 19.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.27127691071330595, "source_start_sample": 3840, "source_end_sample": 938400, "first_word_output_sample": 3840, "last_word_output_sample": 927360, "samples": 934560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.45, "added_silence_s": 0.0, "duration_s": 19.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u00-c04.mp3" }, "de-conversation-03-base17-block-free-u00-c05": { "id": "de-conversation-03-base17-block-free-u00-c05", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 85606, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 919680, "script_start_s": 0.16, "script_end_s": 19.16, "all_asr_last_word_s": 59.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08163265306122448, "source_identity_reward": 0.29974558324931905, "source_start_sample": 3840, "source_end_sample": 926880, "first_word_output_sample": 3840, "last_word_output_sample": 915840, "samples": 923040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.69, "added_silence_s": 0.0, "duration_s": 19.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u00-c05.mp3" }, "de-conversation-03-base17-block-free-u00-c06": { "id": "de-conversation-03-base17-block-free-u00-c06", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 85607, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 19.76, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 940800, "script_start_s": 0.16, "script_end_s": 19.6, "all_asr_last_word_s": 19.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.323971703672796, "source_start_sample": 3840, "source_end_sample": 948000, "first_word_output_sample": 3840, "last_word_output_sample": 936960, "samples": 944160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u00-c06.mp3" }, "de-conversation-03-base17-block-free-u00-c07": { "id": "de-conversation-03-base17-block-free-u00-c07", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 85608, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 21.2, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1002240, "script_start_s": 0.16, "script_end_s": 20.88, "all_asr_last_word_s": 20.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.24253209229722228, "source_start_sample": 3840, "source_end_sample": 1009440, "first_word_output_sample": 3840, "last_word_output_sample": 998400, "samples": 1005600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.17, "added_silence_s": 0.0, "duration_s": 20.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u00-c07.mp3" }, "de-conversation-01-base17-segment-timed-u02-c00": { "id": "de-conversation-01-base17-segment-timed-u02-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83133, "plain": "Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz.", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 203520, "script_start_s": 0.0, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5961183044834826, "source_start_sample": 0, "source_end_sample": 207360, "first_word_output_sample": 0, "last_word_output_sample": 203520, "samples": 210720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-timed-u02-c00.mp3" }, "de-conversation-02-base17-block-free-u01-c00": { "id": "de-conversation-02-base17-block-free-u01-c00", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 84617, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1359360, "script_start_s": 0.16, "script_end_s": 28.32, "all_asr_last_word_s": 28.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.208955223880597, "source_identity_reward": 0.15940599420717572, "source_start_sample": 3840, "source_end_sample": 1366560, "first_word_output_sample": 3840, "last_word_output_sample": 1355520, "samples": 1362720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 31.53, "added_silence_s": 0.0, "duration_s": 28.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u01-c00.mp3" }, "de-conversation-02-base17-block-free-u01-c01": { "id": "de-conversation-02-base17-block-free-u01-c01", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 84618, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1040640, "script_start_s": 0.16, "script_end_s": 21.68, "all_asr_last_word_s": 27.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.23880597014925373, "source_identity_reward": 0.19722723895223798, "source_start_sample": 3840, "source_end_sample": 1047840, "first_word_output_sample": 3840, "last_word_output_sample": 1036800, "samples": 1044000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 38.17, "added_silence_s": 0.0, "duration_s": 21.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u01-c01.mp3" }, "de-conversation-02-base17-block-free-u01-c02": { "id": "de-conversation-02-base17-block-free-u01-c02", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 84619, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 30.4, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1040640, "script_start_s": 0.16, "script_end_s": 21.68, "all_asr_last_word_s": 28.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1791044776119403, "source_identity_reward": 0.19647743744452417, "source_start_sample": 3840, "source_end_sample": 1047840, "first_word_output_sample": 3840, "last_word_output_sample": 1036800, "samples": 1044000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 8.57, "added_silence_s": 0.0, "duration_s": 21.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u01-c02.mp3" }, "de-conversation-02-base17-block-free-u01-c03": { "id": "de-conversation-02-base17-block-free-u01-c03", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 84620, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 67, "matched_tokens": 52, "coverage": 0.7761194029850746, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.22388059701492538, "source_identity_reward": 0.1529505235242474 }, "de-conversation-02-base17-block-free-u01-c04": { "id": "de-conversation-02-base17-block-free-u01-c04", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 84621, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1009920, "script_start_s": 0.0, "script_end_s": 21.04, "all_asr_last_word_s": 23.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04477611940298507, "source_identity_reward": 0.25404451989515275, "source_start_sample": 0, "source_end_sample": 1017120, "first_word_output_sample": 0, "last_word_output_sample": 1009920, "samples": 1017120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.81, "added_silence_s": 0.0, "duration_s": 21.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u01-c04.mp3" }, "de-conversation-02-base17-block-free-u01-c05": { "id": "de-conversation-02-base17-block-free-u01-c05", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 84622, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 28.16, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 64, "coverage": 0.9552238805970149, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1067520, "script_start_s": 0.16, "script_end_s": 22.24, "all_asr_last_word_s": 27.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.208955223880597, "source_identity_reward": 0.16945490066802904, "source_start_sample": 3840, "source_end_sample": 1074720, "first_word_output_sample": 3840, "last_word_output_sample": 1063680, "samples": 1070880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 5.77, "added_silence_s": 0.0, "duration_s": 22.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u01-c05.mp3" }, "de-conversation-02-base17-block-free-u01-c06": { "id": "de-conversation-02-base17-block-free-u01-c06", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 84623, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 27.12, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1036800, "script_start_s": 0.16, "script_end_s": 21.6, "all_asr_last_word_s": 26.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.208955223880597, "source_identity_reward": 0.16774103676139226, "source_start_sample": 3840, "source_end_sample": 1044000, "first_word_output_sample": 3840, "last_word_output_sample": 1032960, "samples": 1040160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 5.37, "added_silence_s": 0.0, "duration_s": 21.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u01-c06.mp3" }, "de-conversation-02-base17-block-free-u01-c07": { "id": "de-conversation-02-base17-block-free-u01-c07", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 84624, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1082880, "script_start_s": 0.16, "script_end_s": 22.56, "all_asr_last_word_s": 28.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.208955223880597, "source_identity_reward": 0.18961650013687917, "source_start_sample": 3840, "source_end_sample": 1090080, "first_word_output_sample": 3840, "last_word_output_sample": 1079040, "samples": 1086240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 37.29, "added_silence_s": 0.0, "duration_s": 22.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u01-c07.mp3" }, "en-conversation-05-base17-block-timed-u01-c00": { "id": "en-conversation-05-base17-block-timed-u01-c00", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 77617, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 936960, "script_start_s": 0.16, "script_end_s": 19.52, "all_asr_last_word_s": 19.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06779661016949153, "source_identity_reward": 0.11273009135155322, "source_start_sample": 3840, "source_end_sample": 944160, "first_word_output_sample": 3840, "last_word_output_sample": 933120, "samples": 940320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u01-c00.mp3" }, "en-conversation-05-base17-block-timed-u01-c01": { "id": "en-conversation-05-base17-block-timed-u01-c01", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 77618, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 936960, "script_start_s": 0.16, "script_end_s": 19.52, "all_asr_last_word_s": 19.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06779661016949153, "source_identity_reward": 0.17960942515256234, "source_start_sample": 3840, "source_end_sample": 944160, "first_word_output_sample": 3840, "last_word_output_sample": 933120, "samples": 940320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u01-c01.mp3" }, "en-conversation-05-base17-block-timed-u01-c02": { "id": "en-conversation-05-base17-block-timed-u01-c02", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 77619, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 936960, "script_start_s": 0.16, "script_end_s": 19.52, "all_asr_last_word_s": 19.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06779661016949153, "source_identity_reward": 0.1675341310974457, "source_start_sample": 3840, "source_end_sample": 944160, "first_word_output_sample": 3840, "last_word_output_sample": 933120, "samples": 940320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u01-c02.mp3" }, "en-conversation-05-base17-block-timed-u01-c03": { "id": "en-conversation-05-base17-block-timed-u01-c03", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 77620, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 936960, "script_start_s": 0.16, "script_end_s": 19.52, "all_asr_last_word_s": 19.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06779661016949153, "source_identity_reward": 0.19167677892329849, "source_start_sample": 3840, "source_end_sample": 944160, "first_word_output_sample": 3840, "last_word_output_sample": 933120, "samples": 940320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u01-c03.mp3" }, "en-conversation-05-base17-block-timed-u01-c04": { "id": "en-conversation-05-base17-block-timed-u01-c04", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 77621, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 56, "coverage": 0.9491525423728814, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 936960, "script_start_s": 0.08, "script_end_s": 19.52, "all_asr_last_word_s": 19.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05084745762711865, "source_identity_reward": 0.10215941065176816, "source_start_sample": 0, "source_end_sample": 944160, "first_word_output_sample": 3840, "last_word_output_sample": 936960, "samples": 944160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u01-c04.mp3" }, "en-conversation-05-base17-block-timed-u01-c05": { "id": "en-conversation-05-base17-block-timed-u01-c05", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 77622, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 936960, "script_start_s": 0.16, "script_end_s": 19.52, "all_asr_last_word_s": 19.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06779661016949153, "source_identity_reward": 0.08848330132033203, "source_start_sample": 3840, "source_end_sample": 944160, "first_word_output_sample": 3840, "last_word_output_sample": 933120, "samples": 940320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u01-c05.mp3" }, "en-conversation-05-base17-block-timed-u01-c06": { "id": "en-conversation-05-base17-block-timed-u01-c06", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 77623, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 54, "coverage": 0.9152542372881356, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 936960, "script_start_s": 0.16, "script_end_s": 19.52, "all_asr_last_word_s": 19.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0847457627118644, "source_identity_reward": 0.14159382235703374, "source_start_sample": 3840, "source_end_sample": 944160, "first_word_output_sample": 3840, "last_word_output_sample": 933120, "samples": 940320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u01-c06.mp3" }, "en-conversation-05-base17-block-timed-u01-c07": { "id": "en-conversation-05-base17-block-timed-u01-c07", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 77624, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 54, "coverage": 0.9152542372881356, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 936960, "script_start_s": 0.16, "script_end_s": 19.52, "all_asr_last_word_s": 19.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0847457627118644, "source_identity_reward": 0.2008371862963594, "source_start_sample": 3840, "source_end_sample": 944160, "first_word_output_sample": 3840, "last_word_output_sample": 933120, "samples": 940320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-timed-u01-c07.mp3" }, "de-conversation-05-base17-block-free-u03-c00": { "id": "de-conversation-05-base17-block-free-u03-c00", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 87649, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 70, "matched_tokens": 68, "coverage": 0.9714285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1006080, "script_start_s": 0.0, "script_end_s": 20.96, "all_asr_last_word_s": 40.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08571428571428572, "source_identity_reward": 0.21947876482128473, "source_start_sample": 0, "source_end_sample": 1013280, "first_word_output_sample": 0, "last_word_output_sample": 1006080, "samples": 1013280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.89, "added_silence_s": 0.0, "duration_s": 21.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u03-c00.mp3" }, "de-conversation-05-base17-block-free-u03-c01": { "id": "de-conversation-05-base17-block-free-u03-c01", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 87650, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 23.52, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 67, "coverage": 0.9571428571428572, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1067520, "script_start_s": 0.16, "script_end_s": 22.24, "all_asr_last_word_s": 23.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u03-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08571428571428572, "source_identity_reward": 0.3010386069474429, "source_start_sample": 3840, "source_end_sample": 1074720, "first_word_output_sample": 3840, "last_word_output_sample": 1063680, "samples": 1070880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.13, "added_silence_s": 0.0, "duration_s": 22.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u03-c01.mp3" }, "de-conversation-05-base17-block-free-u03-c02": { "id": "de-conversation-05-base17-block-free-u03-c02", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 87651, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 29.76, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1132800, "script_start_s": 0.16, "script_end_s": 23.6, "all_asr_last_word_s": 23.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u03-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1, "source_identity_reward": 0.23201251173510204, "source_start_sample": 3840, "source_end_sample": 1140000, "first_word_output_sample": 3840, "last_word_output_sample": 1128960, "samples": 1136160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 6.01, "added_silence_s": 0.0, "duration_s": 23.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u03-c02.mp3" }, "de-conversation-05-base17-block-free-u03-c03": { "id": "de-conversation-05-base17-block-free-u03-c03", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 87652, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.24, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 67, "coverage": 0.9571428571428572, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1094400, "script_start_s": 0.0, "script_end_s": 22.8, "all_asr_last_word_s": 24.080000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u03-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.14285714285714285, "source_identity_reward": 0.23279605226028244, "source_start_sample": 0, "source_end_sample": 1101600, "first_word_output_sample": 0, "last_word_output_sample": 1094400, "samples": 1101600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.29, "added_silence_s": 0.0, "duration_s": 22.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u03-c03.mp3" }, "de-conversation-05-base17-block-free-u03-c04": { "id": "de-conversation-05-base17-block-free-u03-c04", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 87653, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 21.52, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1025280, "script_start_s": 0.08, "script_end_s": 21.36, "all_asr_last_word_s": 21.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u03-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014285714285714285, "source_identity_reward": 0.3530604630730937, "source_start_sample": 0, "source_end_sample": 1032480, "first_word_output_sample": 3840, "last_word_output_sample": 1025280, "samples": 1032480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u03-c04.mp3" }, "de-conversation-05-base17-block-free-u03-c05": { "id": "de-conversation-05-base17-block-free-u03-c05", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 87654, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 70, "matched_tokens": 47, "coverage": 0.6714285714285714, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u03-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.32857142857142857, "source_identity_reward": 0.10219131018086824 }, "de-conversation-05-base17-block-free-u03-c06": { "id": "de-conversation-05-base17-block-free-u03-c06", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 87655, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 70, "matched_tokens": 68, "coverage": 0.9714285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1079040, "script_start_s": 0.0, "script_end_s": 22.48, "all_asr_last_word_s": 38.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u03-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.11428571428571428, "source_identity_reward": 0.24640698475413525, "source_start_sample": 0, "source_end_sample": 1086240, "first_word_output_sample": 0, "last_word_output_sample": 1079040, "samples": 1086240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 37.37, "added_silence_s": 0.0, "duration_s": 22.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u03-c06.mp3" }, "de-conversation-05-base17-block-free-u03-c07": { "id": "de-conversation-05-base17-block-free-u03-c07", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 87656, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 70, "matched_tokens": 68, "coverage": 0.9714285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 979200, "script_start_s": 0.08, "script_end_s": 20.4, "all_asr_last_word_s": 58.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u03-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.12857142857142856, "source_identity_reward": 0.1889936003478166, "source_start_sample": 0, "source_end_sample": 986400, "first_word_output_sample": 3840, "last_word_output_sample": 979200, "samples": 986400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 39.45, "added_silence_s": 0.0, "duration_s": 20.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u03-c07.mp3" }, "de-conversation-03-base17-segment-timed-u11-c00": { "id": "de-conversation-03-base17-segment-timed-u11-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 11, "coverage": 0.8461538461538461, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.15384615384615385, "source_identity_reward": 0.5398429542624916 }, "de-conversation-01-base17-segment-free-u10-c00": { "id": "de-conversation-01-base17-segment-free-u10-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 311040, "script_start_s": 0.0, "script_end_s": 6.48, "all_asr_last_word_s": 6.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.708471280975718, "source_start_sample": 0, "source_end_sample": 318240, "first_word_output_sample": 0, "last_word_output_sample": 311040, "samples": 318240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.37, "added_silence_s": 0.0, "duration_s": 6.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-free-u10-c00.mp3" }, "de-conversation-03-base17-segment-free-u03-c00": { "id": "de-conversation-03-base17-segment-free-u03-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85149, "plain": "Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 9.92, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 291840, "script_start_s": 0.24, "script_end_s": 6.08, "all_asr_last_word_s": 6.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6790218204203737, "source_start_sample": 7680, "source_end_sample": 299040, "first_word_output_sample": 3840, "last_word_output_sample": 284160, "samples": 291360, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 3.69, "added_silence_s": 0.0, "duration_s": 6.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-free-u03-c00.mp3" }, "en-conversation-04-base17-segment-timed-u07-c00": { "id": "en-conversation-04-base17-segment-timed-u07-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76213, "plain": "A reassuringly human approach to a very peculiar little machine.", "source_duration_s": 3.44, "source_frame_cap": false, "alignment": { "expected_tokens": 10, "matched_tokens": 10, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 161280, "script_start_s": 0.16, "script_end_s": 3.36, "all_asr_last_word_s": 3.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.19530314997020146, "source_start_sample": 3840, "source_end_sample": 165120, "first_word_output_sample": 3840, "last_word_output_sample": 157440, "samples": 164640, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-timed-u07-c00.mp3" }, "en-conversation-02-base17-segment-free-u00-c00": { "id": "en-conversation-02-base17-segment-free-u00-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74101, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model.", "source_duration_s": 8.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 272640, "script_start_s": 0.16, "script_end_s": 5.68, "all_asr_last_word_s": 5.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.31958996858721567, "source_start_sample": 3840, "source_end_sample": 279840, "first_word_output_sample": 3840, "last_word_output_sample": 268800, "samples": 276000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 2.89, "added_silence_s": 0.0, "duration_s": 5.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-free-u00-c00.mp3" }, "en-conversation-04-base17-segment-free-u03-c00": { "id": "en-conversation-04-base17-segment-free-u03-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76149, "plain": "Thank you. I was about to ask it to fit my entire life into three seconds.", "source_duration_s": 10.0, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 249600, "script_start_s": 0.16, "script_end_s": 5.2, "all_asr_last_word_s": 5.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.07746135879331091, "source_start_sample": 3840, "source_end_sample": 256800, "first_word_output_sample": 3840, "last_word_output_sample": 245760, "samples": 252960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 4.65, "added_silence_s": 0.0, "duration_s": 5.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-free-u03-c00.mp3" }, "en-conversation-05-base17-segment-free-u05-c00": { "id": "en-conversation-05-base17-segment-free-u05-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77181, "plain": "I want surprise, warmth, a little teasing, and no one reading the stage directions aloud.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 357120, "script_start_s": 0.16, "script_end_s": 7.44, "all_asr_last_word_s": 28.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 2.2, "source_identity_reward": 0.08412761873322863, "source_start_sample": 3840, "source_end_sample": 364320, "first_word_output_sample": 3840, "last_word_output_sample": 353280, "samples": 360480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 22.41, "added_silence_s": 0.0, "duration_s": 7.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-free-u05-c00.mp3" }, "de-conversation-03-podcast-segment-timed-u06-c00": { "id": "de-conversation-03-podcast-segment-timed-u06-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85197, "plain": "Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie.", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 203520, "script_start_s": 0.0, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5286467008339111, "source_start_sample": 0, "source_end_sample": 207360, "first_word_output_sample": 0, "last_word_output_sample": 203520, "samples": 210720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-timed-u06-c00.mp3" }, "en-conversation-01-podcast-block-free-u02-c00": { "id": "en-conversation-01-podcast-block-free-u02-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 73633, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 24.08, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 890880, "script_start_s": 0.16, "script_end_s": 18.56, "all_asr_last_word_s": 24.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2909090909090909, "source_identity_reward": 0.27411333153778783, "source_start_sample": 3840, "source_end_sample": 898080, "first_word_output_sample": 3840, "last_word_output_sample": 887040, "samples": 894240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 5.37, "added_silence_s": 0.0, "duration_s": 18.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u02-c00.mp3" }, "en-conversation-01-podcast-block-free-u02-c01": { "id": "en-conversation-01-podcast-block-free-u02-c01", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 73634, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 19.36, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 921600, "script_start_s": 0.16, "script_end_s": 19.2, "all_asr_last_word_s": 19.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4171601509632505, "source_start_sample": 3840, "source_end_sample": 928800, "first_word_output_sample": 3840, "last_word_output_sample": 917760, "samples": 924960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u02-c01.mp3" }, "en-conversation-01-podcast-block-free-u02-c02": { "id": "en-conversation-01-podcast-block-free-u02-c02", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 73635, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 23.84, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 921600, "script_start_s": 0.0, "script_end_s": 19.2, "all_asr_last_word_s": 21.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05454545454545454, "source_identity_reward": 0.32030187065632315, "source_start_sample": 0, "source_end_sample": 928800, "first_word_output_sample": 0, "last_word_output_sample": 921600, "samples": 928800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.49, "added_silence_s": 0.0, "duration_s": 19.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u02-c02.mp3" }, "en-conversation-01-podcast-block-free-u02-c03": { "id": "en-conversation-01-podcast-block-free-u02-c03", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 73636, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 20.64, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 921600, "script_start_s": 0.0, "script_end_s": 19.2, "all_asr_last_word_s": 19.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5334565378101221, "source_start_sample": 0, "source_end_sample": 928800, "first_word_output_sample": 0, "last_word_output_sample": 921600, "samples": 928800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.29, "added_silence_s": 0.0, "duration_s": 19.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u02-c03.mp3" }, "en-conversation-01-podcast-block-free-u02-c04": { "id": "en-conversation-01-podcast-block-free-u02-c04", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 73637, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 19.84, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 917760, "script_start_s": 0.0, "script_end_s": 19.12, "all_asr_last_word_s": 19.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4252798224863288, "source_start_sample": 0, "source_end_sample": 924960, "first_word_output_sample": 0, "last_word_output_sample": 917760, "samples": 924960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.57, "added_silence_s": 0.0, "duration_s": 19.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u02-c04.mp3" }, "en-conversation-01-podcast-block-free-u02-c05": { "id": "en-conversation-01-podcast-block-free-u02-c05", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 73638, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.88, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 902400, "script_start_s": 0.16, "script_end_s": 18.8, "all_asr_last_word_s": 18.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4476955526774922, "source_start_sample": 3840, "source_end_sample": 906240, "first_word_output_sample": 3840, "last_word_output_sample": 898560, "samples": 905760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u02-c05.mp3" }, "en-conversation-01-podcast-block-free-u02-c06": { "id": "en-conversation-01-podcast-block-free-u02-c06", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 73639, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.88, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 898560, "script_start_s": 0.16, "script_end_s": 18.72, "all_asr_last_word_s": 18.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5137708745847296, "source_start_sample": 3840, "source_end_sample": 905760, "first_word_output_sample": 3840, "last_word_output_sample": 894720, "samples": 901920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 18.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u02-c06.mp3" }, "en-conversation-01-podcast-block-free-u02-c07": { "id": "en-conversation-01-podcast-block-free-u02-c07", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 73640, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 933120, "script_start_s": 0.0, "script_end_s": 19.44, "all_asr_last_word_s": 22.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05454545454545454, "source_identity_reward": 0.40962557907230784, "source_start_sample": 0, "source_end_sample": 940320, "first_word_output_sample": 0, "last_word_output_sample": 933120, "samples": 940320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 40.41, "added_silence_s": 0.0, "duration_s": 19.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u02-c07.mp3" }, "de-conversation-04-podcast-segment-timed-u02-c00": { "id": "de-conversation-04-podcast-segment-timed-u02-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86133, "plain": "Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 11, "coverage": 0.7333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 222720, "script_start_s": 0.08, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.26666666666666666, "source_identity_reward": 0.3333892489965819, "source_start_sample": 0, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-timed-u02-c00.mp3" }, "de-conversation-04-podcast-segment-free-u10-c00": { "id": "de-conversation-04-podcast-segment-free-u10-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 276480, "script_start_s": 0.08, "script_end_s": 5.76, "all_asr_last_word_s": 19.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.25, "source_identity_reward": 0.6806941670883921, "source_start_sample": 0, "source_end_sample": 283680, "first_word_output_sample": 3840, "last_word_output_sample": 276480, "samples": 283680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.09, "added_silence_s": 0.0, "duration_s": 5.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-free-u10-c00.mp3" }, "en-conversation-04-podcast-segment-free-u11-c00": { "id": "en-conversation-04-podcast-segment-free-u11-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 6.4, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 245760, "script_start_s": 0.16, "script_end_s": 5.12, "all_asr_last_word_s": 5.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.19361414152557188, "source_start_sample": 3840, "source_end_sample": 252960, "first_word_output_sample": 3840, "last_word_output_sample": 241920, "samples": 249120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.13, "added_silence_s": 0.0, "duration_s": 5.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-free-u11-c00.mp3" }, "de-conversation-02-podcast-segment-free-u11-c00": { "id": "de-conversation-02-podcast-segment-free-u11-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 21.6, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 311040, "script_start_s": 0.24, "script_end_s": 6.48, "all_asr_last_word_s": 21.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.9230769230769231, "source_identity_reward": 0.352113892070618, "source_start_sample": 7680, "source_end_sample": 318240, "first_word_output_sample": 3840, "last_word_output_sample": 303360, "samples": 310560, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 14.97, "added_silence_s": 0.0, "duration_s": 6.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-free-u11-c00.mp3" }, "de-conversation-03-podcast-segment-free-u11-c00": { "id": "de-conversation-03-podcast-segment-free-u11-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 322560, "script_start_s": 0.24, "script_end_s": 6.72, "all_asr_last_word_s": 6.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5493854921984541, "source_start_sample": 7680, "source_end_sample": 329760, "first_word_output_sample": 3840, "last_word_output_sample": 314880, "samples": 322080, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 23.13, "added_silence_s": 0.0, "duration_s": 6.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-free-u11-c00.mp3" }, "en-conversation-05-podcast-segment-free-u15-c00": { "id": "en-conversation-05-podcast-segment-free-u15-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77341, "plain": "Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 19, "matched_tokens": 19, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 326400, "script_start_s": 0.0, "script_end_s": 6.8, "all_asr_last_word_s": 20.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u15-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3684210526315789, "source_identity_reward": 0.07493943373251809, "source_start_sample": 0, "source_end_sample": 333600, "first_word_output_sample": 0, "last_word_output_sample": 326400, "samples": 333600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.05, "added_silence_s": 0.0, "duration_s": 6.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-free-u15-c00.mp3" }, "de-conversation-01-podcast-block-free-u01-c00": { "id": "de-conversation-01-podcast-block-free-u01-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 83617, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1061760, "script_start_s": 0.16, "script_end_s": 22.12, "all_asr_last_word_s": 22.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.7634647632676932, "source_start_sample": 3840, "source_end_sample": 1068960, "first_word_output_sample": 3840, "last_word_output_sample": 1057920, "samples": 1065120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 37.73, "added_silence_s": 0.0, "duration_s": 22.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u01-c00.mp3" }, "de-conversation-01-podcast-block-free-u01-c01": { "id": "de-conversation-01-podcast-block-free-u01-c01", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 83618, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 24.08, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1059840, "script_start_s": 0.0, "script_end_s": 22.08, "all_asr_last_word_s": 22.080000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.7559687006076252, "source_start_sample": 0, "source_end_sample": 1067040, "first_word_output_sample": 0, "last_word_output_sample": 1059840, "samples": 1067040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.85, "added_silence_s": 0.0, "duration_s": 22.23, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u01-c01.mp3" }, "de-conversation-01-podcast-block-free-u01-c02": { "id": "de-conversation-01-podcast-block-free-u01-c02", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 83619, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1044480, "script_start_s": 0.0, "script_end_s": 21.76, "all_asr_last_word_s": 59.480000000000004 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.52, "source_identity_reward": 0.5565606597477475, "source_start_sample": 0, "source_end_sample": 1051680, "first_word_output_sample": 0, "last_word_output_sample": 1044480, "samples": 1051680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.09, "added_silence_s": 0.0, "duration_s": 21.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u01-c02.mp3" }, "de-conversation-01-podcast-block-free-u01-c03": { "id": "de-conversation-01-podcast-block-free-u01-c03", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 83620, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 20.72, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 986880, "script_start_s": 0.0, "script_end_s": 20.56, "all_asr_last_word_s": 20.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.7932544378720738, "source_start_sample": 0, "source_end_sample": 994080, "first_word_output_sample": 0, "last_word_output_sample": 986880, "samples": 994080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u01-c03.mp3" }, "de-conversation-01-podcast-block-free-u01-c04": { "id": "de-conversation-01-podcast-block-free-u01-c04", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 83621, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 30.64, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 990720, "script_start_s": 0.16, "script_end_s": 20.64, "all_asr_last_word_s": 30.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08, "source_identity_reward": 0.7913274484188926, "source_start_sample": 3840, "source_end_sample": 997920, "first_word_output_sample": 3840, "last_word_output_sample": 986880, "samples": 994080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 9.85, "added_silence_s": 0.0, "duration_s": 20.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u01-c04.mp3" }, "de-conversation-01-podcast-block-free-u01-c05": { "id": "de-conversation-01-podcast-block-free-u01-c05", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 83622, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 35.28, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1032960, "script_start_s": 0.16, "script_end_s": 21.52, "all_asr_last_word_s": 34.879999999999995 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4, "source_identity_reward": 0.5335826650629932, "source_start_sample": 3840, "source_end_sample": 1040160, "first_word_output_sample": 3840, "last_word_output_sample": 1029120, "samples": 1036320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 13.61, "added_silence_s": 0.0, "duration_s": 21.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u01-c05.mp3" }, "de-conversation-01-podcast-block-free-u01-c06": { "id": "de-conversation-01-podcast-block-free-u01-c06", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 83623, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1021440, "script_start_s": 0.0, "script_end_s": 21.28, "all_asr_last_word_s": 47.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0, "source_identity_reward": 0.39221374732793, "source_start_sample": 0, "source_end_sample": 1028640, "first_word_output_sample": 0, "last_word_output_sample": 1021440, "samples": 1028640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.57, "added_silence_s": 0.0, "duration_s": 21.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u01-c06.mp3" }, "de-conversation-01-podcast-block-free-u01-c07": { "id": "de-conversation-01-podcast-block-free-u01-c07", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 83624, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 21.84, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1040640, "script_start_s": 0.16, "script_end_s": 21.68, "all_asr_last_word_s": 21.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.617006578275274, "source_start_sample": 3840, "source_end_sample": 1047840, "first_word_output_sample": 3840, "last_word_output_sample": 1036800, "samples": 1044000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-free-u01-c07.mp3" }, "en-conversation-02-podcast-segment-timed-u06-c00": { "id": "en-conversation-02-podcast-segment-timed-u06-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74197, "plain": "That is why the reference clips sit beside the output. You can compare for yourself.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 234240, "script_start_s": 0.0, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.22502188319577385, "source_start_sample": 0, "source_end_sample": 241440, "first_word_output_sample": 0, "last_word_output_sample": 234240, "samples": 241440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-timed-u06-c00.mp3" }, "en-conversation-01-podcast-block-free-u00-c00": { "id": "en-conversation-01-podcast-block-free-u00-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 73601, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 17.36, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 49, "coverage": 0.9423076923076923, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 825600, "script_start_s": 0.0, "script_end_s": 17.2, "all_asr_last_word_s": 17.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.057692307692307696, "source_identity_reward": 0.5828547526187391, "source_start_sample": 0, "source_end_sample": 832800, "first_word_output_sample": 0, "last_word_output_sample": 825600, "samples": 832800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u00-c00.mp3" }, "en-conversation-01-podcast-block-free-u00-c01": { "id": "en-conversation-01-podcast-block-free-u00-c01", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 73602, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 21.6, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 49, "coverage": 0.9423076923076923, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 833280, "script_start_s": 0.0, "script_end_s": 17.36, "all_asr_last_word_s": 17.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.057692307692307696, "source_identity_reward": 0.5878071694638599, "source_start_sample": 0, "source_end_sample": 840480, "first_word_output_sample": 0, "last_word_output_sample": 833280, "samples": 840480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.09, "added_silence_s": 0.0, "duration_s": 17.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u00-c01.mp3" }, "en-conversation-01-podcast-block-free-u00-c02": { "id": "en-conversation-01-podcast-block-free-u00-c02", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 73603, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 36.56, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 840960, "script_start_s": 0.0, "script_end_s": 17.52, "all_asr_last_word_s": 35.480000000000004 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.8076923076923077, "source_identity_reward": 0.3546245071521994, "source_start_sample": 0, "source_end_sample": 848160, "first_word_output_sample": 0, "last_word_output_sample": 840960, "samples": 848160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 18.89, "added_silence_s": 0.0, "duration_s": 17.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u00-c02.mp3" }, "en-conversation-01-podcast-block-free-u00-c03": { "id": "en-conversation-01-podcast-block-free-u00-c03", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 73604, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 49, "coverage": 0.9423076923076923, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 871680, "script_start_s": 0.0, "script_end_s": 18.16, "all_asr_last_word_s": 18.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.057692307692307696, "source_identity_reward": 0.5321579145153282, "source_start_sample": 0, "source_end_sample": 878880, "first_word_output_sample": 0, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 18.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u00-c03.mp3" }, "en-conversation-01-podcast-block-free-u00-c04": { "id": "en-conversation-01-podcast-block-free-u00-c04", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 73605, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 18.08, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 49, "coverage": 0.9423076923076923, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 814080, "script_start_s": 0.0, "script_end_s": 16.96, "all_asr_last_word_s": 16.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.057692307692307696, "source_identity_reward": 0.5715583095643453, "source_start_sample": 0, "source_end_sample": 821280, "first_word_output_sample": 0, "last_word_output_sample": 814080, "samples": 821280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.97, "added_silence_s": 0.0, "duration_s": 17.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u00-c04.mp3" }, "en-conversation-01-podcast-block-free-u00-c05": { "id": "en-conversation-01-podcast-block-free-u00-c05", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 73606, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 856320, "script_start_s": 0.0, "script_end_s": 17.84, "all_asr_last_word_s": 59.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.6923076923076923, "source_identity_reward": 0.23465424181607075, "source_start_sample": 0, "source_end_sample": 863520, "first_word_output_sample": 0, "last_word_output_sample": 856320, "samples": 863520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 42.01, "added_silence_s": 0.0, "duration_s": 17.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u00-c05.mp3" }, "en-conversation-01-podcast-block-free-u00-c06": { "id": "en-conversation-01-podcast-block-free-u00-c06", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 73607, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 16.72, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 794880, "script_start_s": 0.0, "script_end_s": 16.56, "all_asr_last_word_s": 16.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.5752985217428701, "source_start_sample": 0, "source_end_sample": 802080, "first_word_output_sample": 0, "last_word_output_sample": 794880, "samples": 802080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u00-c06.mp3" }, "en-conversation-01-podcast-block-free-u00-c07": { "id": "en-conversation-01-podcast-block-free-u00-c07", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 73608, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large. You sound suspiciously excited. Did a speech model actually surprise you? It can read a script with different speakers and sentence-level acting directions. Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 17.28, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 821760, "script_start_s": 0.0, "script_end_s": 17.12, "all_asr_last_word_s": 17.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.057692307692307696, "source_identity_reward": 0.5262632641309345, "source_start_sample": 0, "source_end_sample": 828960, "first_word_output_sample": 0, "last_word_output_sample": 821760, "samples": 828960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-free-u00-c07.mp3" }, "de-conversation-03-podcast-segment-timed-u00-c00": { "id": "de-conversation-03-podcast-segment-timed-u00-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85101, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können.", "source_duration_s": 4.56, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 215040, "script_start_s": 0.0, "script_end_s": 4.48, "all_asr_last_word_s": 4.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6738614900523499, "source_start_sample": 0, "source_end_sample": 218880, "first_word_output_sample": 0, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-timed-u00-c00.mp3" }, "de-conversation-05-podcast-segment-free-u10-c00": { "id": "de-conversation-05-podcast-segment-free-u10-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 20.8, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 295680, "script_start_s": 0.0, "script_end_s": 6.16, "all_asr_last_word_s": 14.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.75, "source_identity_reward": 0.43452875884006037, "source_start_sample": 0, "source_end_sample": 302880, "first_word_output_sample": 0, "last_word_output_sample": 295680, "samples": 302880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 14.49, "added_silence_s": 0.0, "duration_s": 6.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u10-c00.mp3" }, "de-conversation-05-podcast-segment-timed-u00-c00": { "id": "de-conversation-05-podcast-segment-timed-u00-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87101, "plain": "Eine Hörerin erlebt erst Englisch, dann Deutsch, Französisch, Spanisch und Niederländisch.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 192000, "script_start_s": 0.0, "script_end_s": 4.0, "all_asr_last_word_s": 4.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6747346642614356, "source_start_sample": 0, "source_end_sample": 195840, "first_word_output_sample": 0, "last_word_output_sample": 192000, "samples": 199200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-timed-u00-c00.mp3" }, "de-conversation-01-podcast-segment-free-u11-c00": { "id": "de-conversation-01-podcast-segment-free-u11-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 9.44, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 291840, "script_start_s": 0.16, "script_end_s": 6.08, "all_asr_last_word_s": 6.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5735804471532788, "source_start_sample": 3840, "source_end_sample": 299040, "first_word_output_sample": 3840, "last_word_output_sample": 288000, "samples": 295200, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 3.21, "added_silence_s": 0.0, "duration_s": 6.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-free-u11-c00.mp3" }, "en-conversation-05-podcast-segment-free-u09-c00": { "id": "en-conversation-05-podcast-segment-free-u09-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 299520, "script_start_s": 0.16, "script_end_s": 6.24, "all_asr_last_word_s": 29.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.75, "source_identity_reward": 0.055069459445609543, "source_start_sample": 3840, "source_end_sample": 306720, "first_word_output_sample": 3840, "last_word_output_sample": 295680, "samples": 302880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.61, "added_silence_s": 0.0, "duration_s": 6.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-free-u09-c00.mp3" }, "en-conversation-03-podcast-segment-free-u03-c00": { "id": "en-conversation-03-podcast-segment-free-u03-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75149, "plain": "I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 249600, "script_start_s": 0.0, "script_end_s": 5.2, "all_asr_last_word_s": 5.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.17183610243427458, "source_start_sample": 0, "source_end_sample": 256800, "first_word_output_sample": 0, "last_word_output_sample": 249600, "samples": 256800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.65, "added_silence_s": 0.0, "duration_s": 5.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-free-u03-c00.mp3" }, "de-conversation-05-podcast-segment-timed-u04-c00": { "id": "de-conversation-05-podcast-segment-timed-u04-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87165, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 238080, "script_start_s": 0.24, "script_end_s": 4.96, "all_asr_last_word_s": 4.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6997697180967883, "source_start_sample": 7680, "source_end_sample": 241920, "first_word_output_sample": 3840, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-timed-u04-c00.mp3" }, "en-conversation-05-podcast-segment-timed-u08-c00": { "id": "en-conversation-05-podcast-segment-timed-u08-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 222720, "script_start_s": 0.16, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.18654692885602175, "source_start_sample": 3840, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-timed-u08-c00.mp3" }, "en-conversation-03-podcast-segment-timed-u07-c00": { "id": "en-conversation-03-podcast-segment-timed-u07-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75213, "plain": "Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 3.84, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 180480, "script_start_s": 0.16, "script_end_s": 3.76, "all_asr_last_word_s": 3.7600000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2674839083472739, "source_start_sample": 3840, "source_end_sample": 184320, "first_word_output_sample": 3840, "last_word_output_sample": 176640, "samples": 183840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-timed-u07-c00.mp3" }, "de-conversation-02-podcast-block-free-u01-c00": { "id": "de-conversation-02-podcast-block-free-u01-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 84617, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 41.84, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1121280, "script_start_s": 0.0, "script_end_s": 23.36, "all_asr_last_word_s": 38.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4925373134328358, "source_identity_reward": 0.49093763853975864, "source_start_sample": 0, "source_end_sample": 1128480, "first_word_output_sample": 0, "last_word_output_sample": 1121280, "samples": 1128480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 18.33, "added_silence_s": 0.0, "duration_s": 23.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u01-c00.mp3" }, "de-conversation-02-podcast-block-free-u01-c01": { "id": "de-conversation-02-podcast-block-free-u01-c01", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 84618, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 24.32, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 67, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1155840, "script_start_s": 0.0, "script_end_s": 24.08, "all_asr_last_word_s": 24.080000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6538548372634853, "source_start_sample": 0, "source_end_sample": 1163040, "first_word_output_sample": 0, "last_word_output_sample": 1155840, "samples": 1163040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 24.23, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u01-c01.mp3" }, "de-conversation-02-podcast-block-free-u01-c02": { "id": "de-conversation-02-podcast-block-free-u01-c02", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 84619, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.76, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1132800, "script_start_s": 0.0, "script_end_s": 23.6, "all_asr_last_word_s": 23.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029850746268656716, "source_identity_reward": 0.6432171824095891, "source_start_sample": 0, "source_end_sample": 1140000, "first_word_output_sample": 0, "last_word_output_sample": 1132800, "samples": 1140000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u01-c02.mp3" }, "de-conversation-02-podcast-block-free-u01-c03": { "id": "de-conversation-02-podcast-block-free-u01-c03", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 84620, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 24.0, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1144320, "script_start_s": 0.0, "script_end_s": 23.84, "all_asr_last_word_s": 23.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.7256752209806484, "source_start_sample": 0, "source_end_sample": 1151520, "first_word_output_sample": 0, "last_word_output_sample": 1144320, "samples": 1151520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u01-c03.mp3" }, "de-conversation-02-podcast-block-free-u01-c04": { "id": "de-conversation-02-podcast-block-free-u01-c04", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 84621, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 24.48, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1167360, "script_start_s": 0.16, "script_end_s": 24.32, "all_asr_last_word_s": 24.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.6735552790430427, "source_start_sample": 3840, "source_end_sample": 1174560, "first_word_output_sample": 3840, "last_word_output_sample": 1163520, "samples": 1170720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 24.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u01-c04.mp3" }, "de-conversation-02-podcast-block-free-u01-c05": { "id": "de-conversation-02-podcast-block-free-u01-c05", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 84622, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 31.12, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1167360, "script_start_s": 0.0, "script_end_s": 24.32, "all_asr_last_word_s": 30.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.23880597014925373, "source_identity_reward": 0.5958654855483219, "source_start_sample": 0, "source_end_sample": 1174560, "first_word_output_sample": 0, "last_word_output_sample": 1167360, "samples": 1174560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 6.65, "added_silence_s": 0.0, "duration_s": 24.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u01-c05.mp3" }, "de-conversation-02-podcast-block-free-u01-c06": { "id": "de-conversation-02-podcast-block-free-u01-c06", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 84623, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 29.44, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1152000, "script_start_s": 0.0, "script_end_s": 24.0, "all_asr_last_word_s": 29.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.22388059701492538, "source_identity_reward": 0.5407186025109199, "source_start_sample": 0, "source_end_sample": 1159200, "first_word_output_sample": 0, "last_word_output_sample": 1152000, "samples": 1159200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 5.29, "added_silence_s": 0.0, "duration_s": 24.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u01-c06.mp3" }, "de-conversation-02-podcast-block-free-u01-c07": { "id": "de-conversation-02-podcast-block-free-u01-c07", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 84624, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.92, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1140480, "script_start_s": 0.0, "script_end_s": 23.76, "all_asr_last_word_s": 23.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029850746268656716, "source_identity_reward": 0.6949204550790125, "source_start_sample": 0, "source_end_sample": 1147680, "first_word_output_sample": 0, "last_word_output_sample": 1140480, "samples": 1147680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u01-c07.mp3" }, "de-conversation-03-podcast-segment-free-u03-c00": { "id": "de-conversation-03-podcast-segment-free-u03-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85149, "plain": "Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 10.72, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 291840, "script_start_s": 0.16, "script_end_s": 6.08, "all_asr_last_word_s": 6.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.617066724103302, "source_start_sample": 3840, "source_end_sample": 299040, "first_word_output_sample": 3840, "last_word_output_sample": 288000, "samples": 295200, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 4.49, "added_silence_s": 0.0, "duration_s": 6.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-free-u03-c00.mp3" }, "en-conversation-02-podcast-segment-free-u02-c00": { "id": "en-conversation-02-podcast-segment-free-u02-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74133, "plain": "Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 12, "coverage": 0.9230769230769231, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 288000, "script_start_s": 0.08, "script_end_s": 6.0, "all_asr_last_word_s": 6.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07692307692307693, "source_identity_reward": 0.3368798074433261, "source_start_sample": 0, "source_end_sample": 295200, "first_word_output_sample": 3840, "last_word_output_sample": 288000, "samples": 295200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.85, "added_silence_s": 0.0, "duration_s": 6.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-free-u02-c00.mp3" }, "de-conversation-02-podcast-segment-timed-u05-c00": { "id": "de-conversation-02-podcast-segment-timed-u05-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84181, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören.", "source_duration_s": 5.36, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 249600, "script_start_s": 0.0, "script_end_s": 5.2, "all_asr_last_word_s": 5.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7286406883367189, "source_start_sample": 0, "source_end_sample": 256800, "first_word_output_sample": 0, "last_word_output_sample": 249600, "samples": 256800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-timed-u05-c00.mp3" }, "de-conversation-05-podcast-segment-timed-u12-c00": { "id": "de-conversation-05-podcast-segment-timed-u12-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87293, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen.", "source_duration_s": 5.44, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u12-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06666666666666667, "source_identity_reward": 0.4225472694314367 }, "en-conversation-03-podcast-segment-free-u10-c00": { "id": "en-conversation-03-podcast-segment-free-u10-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 22.32, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 276480, "script_start_s": 0.0, "script_end_s": 5.76, "all_asr_last_word_s": 22.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.1428571428571428, "source_identity_reward": 0.17623526586798138, "source_start_sample": 0, "source_end_sample": 283680, "first_word_output_sample": 0, "last_word_output_sample": 276480, "samples": 283680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 16.41, "added_silence_s": 0.0, "duration_s": 5.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-free-u10-c00.mp3" }, "en-conversation-04-podcast-block-free-u00-c00": { "id": "en-conversation-04-podcast-block-free-u00-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 76601, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 24.24, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1094400, "script_start_s": 0.0, "script_end_s": 22.8, "all_asr_last_word_s": 24.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04411764705882353, "source_identity_reward": 0.49405496139381844, "source_start_sample": 0, "source_end_sample": 1101600, "first_word_output_sample": 0, "last_word_output_sample": 1094400, "samples": 1101600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.29, "added_silence_s": 0.0, "duration_s": 22.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u00-c00.mp3" }, "en-conversation-04-podcast-block-free-u00-c01": { "id": "en-conversation-04-podcast-block-free-u00-c01", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 76602, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.0, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1048320, "script_start_s": 0.0, "script_end_s": 21.84, "all_asr_last_word_s": 21.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.34071853992499423, "source_start_sample": 0, "source_end_sample": 1055520, "first_word_output_sample": 0, "last_word_output_sample": 1048320, "samples": 1055520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u00-c01.mp3" }, "en-conversation-04-podcast-block-free-u00-c02": { "id": "en-conversation-04-podcast-block-free-u00-c02", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 76603, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 39.2, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1079040, "script_start_s": 0.0, "script_end_s": 22.48, "all_asr_last_word_s": 36.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4264705882352941, "source_identity_reward": 0.408089624149632, "source_start_sample": 0, "source_end_sample": 1086240, "first_word_output_sample": 0, "last_word_output_sample": 1079040, "samples": 1086240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 16.57, "added_silence_s": 0.0, "duration_s": 22.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u00-c02.mp3" }, "en-conversation-04-podcast-block-free-u00-c03": { "id": "en-conversation-04-podcast-block-free-u00-c03", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 76604, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.64, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1079040, "script_start_s": 0.0, "script_end_s": 22.48, "all_asr_last_word_s": 22.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.3435899965450182, "source_start_sample": 0, "source_end_sample": 1086240, "first_word_output_sample": 0, "last_word_output_sample": 1079040, "samples": 1086240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u00-c03.mp3" }, "en-conversation-04-podcast-block-free-u00-c04": { "id": "en-conversation-04-podcast-block-free-u00-c04", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 76605, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.64, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1079040, "script_start_s": 0.0, "script_end_s": 22.48, "all_asr_last_word_s": 22.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.4784586007884797, "source_start_sample": 0, "source_end_sample": 1086240, "first_word_output_sample": 0, "last_word_output_sample": 1079040, "samples": 1086240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u00-c04.mp3" }, "en-conversation-04-podcast-block-free-u00-c05": { "id": "en-conversation-04-podcast-block-free-u00-c05", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 76606, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.56, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 66, "coverage": 0.9705882352941176, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1075200, "script_start_s": 0.0, "script_end_s": 22.4, "all_asr_last_word_s": 22.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029411764705882353, "source_identity_reward": 0.4817985016439239, "source_start_sample": 0, "source_end_sample": 1082400, "first_word_output_sample": 0, "last_word_output_sample": 1075200, "samples": 1082400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u00-c05.mp3" }, "en-conversation-04-podcast-block-free-u00-c06": { "id": "en-conversation-04-podcast-block-free-u00-c06", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 76607, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.48, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1071360, "script_start_s": 0.0, "script_end_s": 22.32, "all_asr_last_word_s": 22.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.5852861927907829, "source_start_sample": 0, "source_end_sample": 1078560, "first_word_output_sample": 0, "last_word_output_sample": 1071360, "samples": 1078560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u00-c06.mp3" }, "en-conversation-04-podcast-block-free-u00-c07": { "id": "en-conversation-04-podcast-block-free-u00-c07", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 76608, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 22.4, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 66, "coverage": 0.9705882352941176, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1067520, "script_start_s": 0.0, "script_end_s": 22.24, "all_asr_last_word_s": 22.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029411764705882353, "source_identity_reward": 0.5055127573285249, "source_start_sample": 0, "source_end_sample": 1074720, "first_word_output_sample": 0, "last_word_output_sample": 1067520, "samples": 1074720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u00-c07.mp3" }, "en-conversation-01-podcast-segment-timed-u10-c00": { "id": "en-conversation-01-podcast-segment-timed-u10-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 4.8, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 226560, "script_start_s": 0.16, "script_end_s": 4.72, "all_asr_last_word_s": 4.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.39156645657738576, "source_start_sample": 3840, "source_end_sample": 230400, "first_word_output_sample": 3840, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-timed-u10-c00.mp3" }, "de-conversation-01-podcast-segment-free-u03-c00": { "id": "de-conversation-01-podcast-segment-free-u03-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83149, "plain": "Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 9.28, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 203520, "script_start_s": 0.16, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3456244440919399, "source_start_sample": 3840, "source_end_sample": 210720, "first_word_output_sample": 3840, "last_word_output_sample": 199680, "samples": 206880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 4.89, "added_silence_s": 0.0, "duration_s": 4.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-free-u03-c00.mp3" }, "en-conversation-02-podcast-segment-timed-u00-c00": { "id": "en-conversation-02-podcast-segment-timed-u00-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74101, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 222720, "script_start_s": 0.16, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.27803720394080517, "source_start_sample": 3840, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-timed-u00-c00.mp3" }, "de-conversation-04-podcast-segment-free-u11-c00": { "id": "de-conversation-04-podcast-segment-free-u11-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 9.52, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 303360, "script_start_s": 0.16, "script_end_s": 6.32, "all_asr_last_word_s": 6.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2692162424182787, "source_start_sample": 3840, "source_end_sample": 310560, "first_word_output_sample": 3840, "last_word_output_sample": 299520, "samples": 306720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 3.05, "added_silence_s": 0.0, "duration_s": 6.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-free-u11-c00.mp3" }, "de-conversation-04-podcast-segment-free-u09-c00": { "id": "de-conversation-04-podcast-segment-free-u09-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 8.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 276480, "script_start_s": 0.16, "script_end_s": 5.76, "all_asr_last_word_s": 5.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.48173882309394267, "source_start_sample": 3840, "source_end_sample": 283680, "first_word_output_sample": 3840, "last_word_output_sample": 272640, "samples": 279840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 2.73, "added_silence_s": 0.0, "duration_s": 5.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-free-u09-c00.mp3" }, "en-conversation-01-podcast-segment-free-u02-c00": { "id": "en-conversation-01-podcast-segment-free-u02-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73133, "plain": "It can read a script with different speakers and sentence-level acting directions.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 334080, "script_start_s": 0.0, "script_end_s": 6.96, "all_asr_last_word_s": 29.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 2.5384615384615383, "source_identity_reward": 0.11355098354989156, "source_start_sample": 0, "source_end_sample": 341280, "first_word_output_sample": 0, "last_word_output_sample": 334080, "samples": 341280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 22.89, "added_silence_s": 0.0, "duration_s": 7.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-free-u02-c00.mp3" }, "en-conversation-05-podcast-segment-free-u05-c00": { "id": "en-conversation-05-podcast-segment-free-u05-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77181, "plain": "I want surprise, warmth, a little teasing, and no one reading the stage directions aloud.", "source_duration_s": 15.76, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 503040, "script_start_s": 0.16, "script_end_s": 10.48, "all_asr_last_word_s": 15.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0, "source_identity_reward": 0.05424521794781675, "source_start_sample": 3840, "source_end_sample": 510240, "first_word_output_sample": 3840, "last_word_output_sample": 499200, "samples": 506400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 5.13, "added_silence_s": 0.0, "duration_s": 10.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-free-u05-c00.mp3" }, "en-conversation-01-podcast-segment-timed-u09-c00": { "id": "en-conversation-01-podcast-segment-timed-u09-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 4.16, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 195840, "script_start_s": 0.16, "script_end_s": 4.08, "all_asr_last_word_s": 4.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.12944433091052945, "source_start_sample": 3840, "source_end_sample": 199680, "first_word_output_sample": 3840, "last_word_output_sample": 192000, "samples": 199200, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-timed-u09-c00.mp3" }, "de-conversation-03-podcast-segment-timed-u02-c00": { "id": "de-conversation-03-podcast-segment-timed-u02-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85133, "plain": "Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf.", "source_duration_s": 4.48, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 211200, "script_start_s": 0.0, "script_end_s": 4.4, "all_asr_last_word_s": 4.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8149638092279821, "source_start_sample": 0, "source_end_sample": 215040, "first_word_output_sample": 0, "last_word_output_sample": 211200, "samples": 218400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-timed-u02-c00.mp3" }, "de-conversation-04-podcast-block-free-u00-c00": { "id": "de-conversation-04-podcast-block-free-u00-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 86601, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 23.52, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 60, "coverage": 0.9523809523809523, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1121280, "script_start_s": 0.0, "script_end_s": 23.36, "all_asr_last_word_s": 23.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.047619047619047616, "source_identity_reward": 0.7901066412096978, "source_start_sample": 0, "source_end_sample": 1128480, "first_word_output_sample": 0, "last_word_output_sample": 1121280, "samples": 1128480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u00-c00.mp3" }, "de-conversation-04-podcast-block-free-u00-c01": { "id": "de-conversation-04-podcast-block-free-u00-c01", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 86602, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 23.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 61, "coverage": 0.9682539682539683, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1132800, "script_start_s": 0.0, "script_end_s": 23.6, "all_asr_last_word_s": 23.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.031746031746031744, "source_identity_reward": 0.8091041932977174, "source_start_sample": 0, "source_end_sample": 1140000, "first_word_output_sample": 0, "last_word_output_sample": 1132800, "samples": 1140000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u00-c01.mp3" }, "de-conversation-04-podcast-block-free-u00-c02": { "id": "de-conversation-04-podcast-block-free-u00-c02", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 86603, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 8.8, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 23, "coverage": 0.36507936507936506, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.6349206349206349, "source_identity_reward": 0.2124561702089576 }, "de-conversation-04-podcast-block-free-u00-c03": { "id": "de-conversation-04-podcast-block-free-u00-c03", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 86604, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 27.84, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 59, "coverage": 0.9365079365079365, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1201920, "script_start_s": 0.0, "script_end_s": 25.04, "all_asr_last_word_s": 26.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1111111111111111, "source_identity_reward": 0.6744291843646478, "source_start_sample": 0, "source_end_sample": 1209120, "first_word_output_sample": 0, "last_word_output_sample": 1201920, "samples": 1209120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.65, "added_silence_s": 0.0, "duration_s": 25.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u00-c03.mp3" }, "de-conversation-04-podcast-block-free-u00-c04": { "id": "de-conversation-04-podcast-block-free-u00-c04", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 86605, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 63, "matched_tokens": 59, "coverage": 0.9365079365079365, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1190400, "script_start_s": 0.0, "script_end_s": 24.8, "all_asr_last_word_s": 59.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.42857142857142855, "source_identity_reward": 0.5769978648419691, "source_start_sample": 0, "source_end_sample": 1197600, "first_word_output_sample": 0, "last_word_output_sample": 1190400, "samples": 1197600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 35.05, "added_silence_s": 0.0, "duration_s": 24.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u00-c04.mp3" }, "de-conversation-04-podcast-block-free-u00-c05": { "id": "de-conversation-04-podcast-block-free-u00-c05", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 86606, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 63, "matched_tokens": 58, "coverage": 0.9206349206349206, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1463040, "script_start_s": 0.0, "script_end_s": 30.48, "all_asr_last_word_s": 30.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2857142857142857, "source_identity_reward": 0.6051126937898063, "source_start_sample": 0, "source_end_sample": 1470240, "first_word_output_sample": 0, "last_word_output_sample": 1463040, "samples": 1470240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 29.37, "added_silence_s": 0.0, "duration_s": 30.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u00-c05.mp3" }, "de-conversation-04-podcast-block-free-u00-c06": { "id": "de-conversation-04-podcast-block-free-u00-c06", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 86607, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 27.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 58, "coverage": 0.9206349206349206, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1167360, "script_start_s": 0.0, "script_end_s": 24.32, "all_asr_last_word_s": 24.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07936507936507936, "source_identity_reward": 0.7356003464725712, "source_start_sample": 0, "source_end_sample": 1174560, "first_word_output_sample": 0, "last_word_output_sample": 1167360, "samples": 1174560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 3.29, "added_silence_s": 0.0, "duration_s": 24.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u00-c06.mp3" }, "de-conversation-04-podcast-block-free-u00-c07": { "id": "de-conversation-04-podcast-block-free-u00-c07", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 86608, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 24.0, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 61, "coverage": 0.9682539682539683, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1144320, "script_start_s": 0.0, "script_end_s": 23.84, "all_asr_last_word_s": 23.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.031746031746031744, "source_identity_reward": 0.7775476204285154, "source_start_sample": 0, "source_end_sample": 1151520, "first_word_output_sample": 0, "last_word_output_sample": 1144320, "samples": 1151520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u00-c07.mp3" }, "de-conversation-05-podcast-block-timed-u02-c00": { "id": "de-conversation-05-podcast-block-timed-u02-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 87633, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.08, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7559740221345932, "source_start_sample": 0, "source_end_sample": 963360, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u02-c00.mp3" }, "de-conversation-05-podcast-block-timed-u02-c01": { "id": "de-conversation-05-podcast-block-timed-u02-c01", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 87634, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.808852356868691, "source_start_sample": 0, "source_end_sample": 960000, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u02-c01.mp3" }, "de-conversation-05-podcast-block-timed-u02-c02": { "id": "de-conversation-05-podcast-block-timed-u02-c02", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 87635, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8117382824493322, "source_start_sample": 0, "source_end_sample": 960000, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u02-c02.mp3" }, "de-conversation-05-podcast-block-timed-u02-c03": { "id": "de-conversation-05-podcast-block-timed-u02-c03", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 87636, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 54, "coverage": 0.9642857142857143, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.7415883646749041 }, "de-conversation-05-podcast-block-timed-u02-c04": { "id": "de-conversation-05-podcast-block-timed-u02-c04", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 87637, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.7632867446253156, "source_start_sample": 0, "source_end_sample": 960000, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u02-c04.mp3" }, "de-conversation-05-podcast-block-timed-u02-c05": { "id": "de-conversation-05-podcast-block-timed-u02-c05", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 87638, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.017857142857142856, "source_identity_reward": 0.6726031413940194, "source_start_sample": 0, "source_end_sample": 960000, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u02-c05.mp3" }, "de-conversation-05-podcast-block-timed-u02-c06": { "id": "de-conversation-05-podcast-block-timed-u02-c06", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 87639, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.08, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 960000, "script_start_s": 0.0, "script_end_s": 20.0, "all_asr_last_word_s": 20.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7807929794444339, "source_start_sample": 0, "source_end_sample": 963840, "first_word_output_sample": 0, "last_word_output_sample": 960000, "samples": 967200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u02-c06.mp3" }, "de-conversation-05-podcast-block-timed-u02-c07": { "id": "de-conversation-05-podcast-block-timed-u02-c07", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 87640, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 54, "coverage": 0.9642857142857143, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.7248909063546625 }, "en-conversation-05-podcast-block-free-u01-c00": { "id": "en-conversation-05-podcast-block-free-u01-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 77617, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.12, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 910080, "script_start_s": 0.0, "script_end_s": 18.96, "all_asr_last_word_s": 18.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06779661016949153, "source_identity_reward": 0.43392936054935094, "source_start_sample": 0, "source_end_sample": 917280, "first_word_output_sample": 0, "last_word_output_sample": 910080, "samples": 917280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u01-c00.mp3" }, "en-conversation-05-podcast-block-free-u01-c01": { "id": "en-conversation-05-podcast-block-free-u01-c01", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 77618, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.52, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06779661016949153, "source_identity_reward": 0.49766249168939397 }, "en-conversation-05-podcast-block-free-u01-c02": { "id": "en-conversation-05-podcast-block-free-u01-c02", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 77619, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 20.56, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 963840, "script_start_s": 0.0, "script_end_s": 20.08, "all_asr_last_word_s": 20.080000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06779661016949153, "source_identity_reward": 0.5953320377629798, "source_start_sample": 0, "source_end_sample": 971040, "first_word_output_sample": 0, "last_word_output_sample": 963840, "samples": 971040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.33, "added_silence_s": 0.0, "duration_s": 20.23, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u01-c02.mp3" }, "en-conversation-05-podcast-block-free-u01-c03": { "id": "en-conversation-05-podcast-block-free-u01-c03", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 77620, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.52, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 56, "coverage": 0.9491525423728814, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 929280, "script_start_s": 0.0, "script_end_s": 19.36, "all_asr_last_word_s": 19.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05084745762711865, "source_identity_reward": 0.5524352355600258, "source_start_sample": 0, "source_end_sample": 936480, "first_word_output_sample": 0, "last_word_output_sample": 929280, "samples": 936480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u01-c03.mp3" }, "en-conversation-05-podcast-block-free-u01-c04": { "id": "en-conversation-05-podcast-block-free-u01-c04", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 77621, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 20.08, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06779661016949153, "source_identity_reward": 0.46059121149204907, "source_start_sample": 0, "source_end_sample": 963360, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u01-c04.mp3" }, "en-conversation-05-podcast-block-free-u01-c05": { "id": "en-conversation-05-podcast-block-free-u01-c05", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 77622, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 20.32, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 57, "coverage": 0.9661016949152542, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 963840, "script_start_s": 0.08, "script_end_s": 20.08, "all_asr_last_word_s": 20.080000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03389830508474576, "source_identity_reward": 0.6004171047057812, "source_start_sample": 0, "source_end_sample": 971040, "first_word_output_sample": 3840, "last_word_output_sample": 963840, "samples": 971040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 20.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u01-c05.mp3" }, "en-conversation-05-podcast-block-free-u01-c06": { "id": "en-conversation-05-podcast-block-free-u01-c06", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 77623, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 20.16, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 960000, "script_start_s": 0.08, "script_end_s": 20.0, "all_asr_last_word_s": 20.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06779661016949153, "source_identity_reward": 0.5048827232895667, "source_start_sample": 0, "source_end_sample": 967200, "first_word_output_sample": 3840, "last_word_output_sample": 960000, "samples": 967200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u01-c06.mp3" }, "en-conversation-05-podcast-block-free-u01-c07": { "id": "en-conversation-05-podcast-block-free-u01-c07", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 77624, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers. I want surprise, warmth, a little teasing, and no one reading the stage directions aloud. Exactly. We publish the full scripts and seeds so the exciting failures remain visible too. All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 19.68, "source_frame_cap": false, "alignment": { "expected_tokens": 59, "matched_tokens": 55, "coverage": 0.9322033898305084, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 917760, "script_start_s": 0.0, "script_end_s": 19.12, "all_asr_last_word_s": 19.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06779661016949153, "source_identity_reward": 0.5888358252926719, "source_start_sample": 0, "source_end_sample": 924960, "first_word_output_sample": 0, "last_word_output_sample": 917760, "samples": 924960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.41, "added_silence_s": 0.0, "duration_s": 19.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u01-c07.mp3" }, "de-conversation-05-podcast-segment-timed-u02-c00": { "id": "de-conversation-05-podcast-segment-timed-u02-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87133, "plain": "Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen.", "source_duration_s": 3.68, "source_frame_cap": false, "alignment": { "expected_tokens": 10, "matched_tokens": 9, "coverage": 0.9, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 172800, "script_start_s": 0.0, "script_end_s": 3.6, "all_asr_last_word_s": 3.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1, "source_identity_reward": 0.685034566784047, "source_start_sample": 0, "source_end_sample": 176640, "first_word_output_sample": 0, "last_word_output_sample": 172800, "samples": 180000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-timed-u02-c00.mp3" }, "en-conversation-05-podcast-segment-free-u12-c00": { "id": "en-conversation-05-podcast-segment-free-u12-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77293, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 341760, "script_start_s": 0.24, "script_end_s": 7.12, "all_asr_last_word_s": 29.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u12-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 2.3333333333333335, "source_identity_reward": 0.13303968069719838, "source_start_sample": 7680, "source_end_sample": 348960, "first_word_output_sample": 3840, "last_word_output_sample": 334080, "samples": 341280, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 22.73, "added_silence_s": 0.0, "duration_s": 7.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-free-u12-c00.mp3" }, "en-conversation-03-podcast-segment-timed-u01-c00": { "id": "en-conversation-03-podcast-segment-timed-u01-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75117, "plain": "Then keeping those voices distinct while the conversation wanders somewhere unexpected?", "source_duration_s": 3.84, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 180480, "script_start_s": 0.16, "script_end_s": 3.76, "all_asr_last_word_s": 3.7600000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.27959302263764874, "source_start_sample": 3840, "source_end_sample": 184320, "first_word_output_sample": 3840, "last_word_output_sample": 176640, "samples": 183840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-timed-u01-c00.mp3" }, "en-conversation-01-podcast-segment-timed-u07-c00": { "id": "en-conversation-01-podcast-segment-timed-u07-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73213, "plain": "All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 222720, "script_start_s": 0.16, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2622139586027531, "source_start_sample": 3840, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-timed-u07-c00.mp3" }, "de-conversation-05-podcast-block-timed-u01-c00": { "id": "de-conversation-05-podcast-block-timed-u01-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 87617, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.2, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 913920, "script_start_s": 0.16, "script_end_s": 19.04, "all_asr_last_word_s": 19.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6685192279533876, "source_start_sample": 3840, "source_end_sample": 921120, "first_word_output_sample": 3840, "last_word_output_sample": 910080, "samples": 917280, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u01-c00.mp3" }, "de-conversation-05-podcast-block-timed-u01-c01": { "id": "de-conversation-05-podcast-block-timed-u01-c01", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 87618, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 917760, "script_start_s": 0.08, "script_end_s": 19.12, "all_asr_last_word_s": 19.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8349247174589272, "source_start_sample": 0, "source_end_sample": 924960, "first_word_output_sample": 3840, "last_word_output_sample": 917760, "samples": 924960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u01-c01.mp3" }, "de-conversation-05-podcast-block-timed-u01-c02": { "id": "de-conversation-05-podcast-block-timed-u01-c02", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 87619, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 53, "coverage": 0.9814814814814815, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 917760, "script_start_s": 0.08, "script_end_s": 19.12, "all_asr_last_word_s": 19.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.018518518518518517, "source_identity_reward": 0.7557127802584512, "source_start_sample": 0, "source_end_sample": 924960, "first_word_output_sample": 3840, "last_word_output_sample": 917760, "samples": 924960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u01-c02.mp3" }, "de-conversation-05-podcast-block-timed-u01-c03": { "id": "de-conversation-05-podcast-block-timed-u01-c03", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 87620, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 53, "coverage": 0.9814814814814815, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 913920, "script_start_s": 0.16, "script_end_s": 19.04, "all_asr_last_word_s": 19.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.018518518518518517, "source_identity_reward": 0.6459451498758042, "source_start_sample": 3840, "source_end_sample": 921120, "first_word_output_sample": 3840, "last_word_output_sample": 910080, "samples": 917280, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 19.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u01-c03.mp3" }, "de-conversation-05-podcast-block-timed-u01-c04": { "id": "de-conversation-05-podcast-block-timed-u01-c04", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 87621, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 53, "coverage": 0.9814814814814815, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 913920, "script_start_s": 0.08, "script_end_s": 19.04, "all_asr_last_word_s": 19.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.018518518518518517, "source_identity_reward": 0.7234914989025205, "source_start_sample": 0, "source_end_sample": 921120, "first_word_output_sample": 3840, "last_word_output_sample": 913920, "samples": 921120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 19.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u01-c04.mp3" }, "de-conversation-05-podcast-block-timed-u01-c05": { "id": "de-conversation-05-podcast-block-timed-u01-c05", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 87622, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 53, "coverage": 0.9814814814814815, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 913920, "script_start_s": 0.16, "script_end_s": 19.04, "all_asr_last_word_s": 19.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.018518518518518517, "source_identity_reward": 0.7442970873134013, "source_start_sample": 3840, "source_end_sample": 921120, "first_word_output_sample": 3840, "last_word_output_sample": 910080, "samples": 917280, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 19.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u01-c05.mp3" }, "de-conversation-05-podcast-block-timed-u01-c06": { "id": "de-conversation-05-podcast-block-timed-u01-c06", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 87623, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 917760, "script_start_s": 0.08, "script_end_s": 19.12, "all_asr_last_word_s": 19.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7858914518383863, "source_start_sample": 0, "source_end_sample": 924960, "first_word_output_sample": 3840, "last_word_output_sample": 917760, "samples": 924960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u01-c06.mp3" }, "de-conversation-05-podcast-block-timed-u01-c07": { "id": "de-conversation-05-podcast-block-timed-u01-c07", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 87624, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 913920, "script_start_s": 0.08, "script_end_s": 19.04, "all_asr_last_word_s": 19.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8050385024813982, "source_start_sample": 0, "source_end_sample": 921120, "first_word_output_sample": 3840, "last_word_output_sample": 913920, "samples": 921120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 19.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u01-c07.mp3" }, "en-conversation-02-podcast-segment-free-u08-c00": { "id": "en-conversation-02-podcast-segment-free-u08-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 8.08, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 276480, "script_start_s": 0.16, "script_end_s": 5.76, "all_asr_last_word_s": 7.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2857142857142857, "source_identity_reward": 0.3175130637994256, "source_start_sample": 3840, "source_end_sample": 283680, "first_word_output_sample": 3840, "last_word_output_sample": 272640, "samples": 279840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 2.17, "added_silence_s": 0.0, "duration_s": 5.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-free-u08-c00.mp3" }, "de-conversation-01-podcast-segment-free-u00-c00": { "id": "de-conversation-01-podcast-segment-free-u00-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83101, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 11, "matched_tokens": 9, "coverage": 0.8181818181818182, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 311040, "script_start_s": 0.0, "script_end_s": 6.48, "all_asr_last_word_s": 29.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.7272727272727273, "source_identity_reward": 0.1863464136818018, "source_start_sample": 0, "source_end_sample": 318240, "first_word_output_sample": 0, "last_word_output_sample": 311040, "samples": 318240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.37, "added_silence_s": 0.0, "duration_s": 6.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-free-u00-c00.mp3" }, "de-conversation-05-podcast-segment-free-u01-c00": { "id": "de-conversation-05-podcast-segment-free-u01-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87117, "plain": "Dieselbe Modellfamilie, andere Stimmen und in jeder Sprache ein eigener emotionaler Rhythmus?", "source_duration_s": 13.52, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 318720, "script_start_s": 0.16, "script_end_s": 6.64, "all_asr_last_word_s": 13.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0, "source_identity_reward": 0.3099799662892143, "source_start_sample": 3840, "source_end_sample": 325920, "first_word_output_sample": 3840, "last_word_output_sample": 314880, "samples": 322080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 6.73, "added_silence_s": 0.0, "duration_s": 6.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u01-c00.mp3" }, "de-conversation-03-podcast-block-free-u00-c00": { "id": "de-conversation-03-podcast-block-free-u00-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 85601, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 37.68, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 998400, "script_start_s": 0.0, "script_end_s": 20.8, "all_asr_last_word_s": 36.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.6530612244897959, "source_identity_reward": 0.5047604797120251, "source_start_sample": 0, "source_end_sample": 1005600, "first_word_output_sample": 0, "last_word_output_sample": 998400, "samples": 1005600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 16.73, "added_silence_s": 0.0, "duration_s": 20.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u00-c00.mp3" }, "de-conversation-03-podcast-block-free-u00-c01": { "id": "de-conversation-03-podcast-block-free-u00-c01", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 85602, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 20.72, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 986880, "script_start_s": 0.0, "script_end_s": 20.56, "all_asr_last_word_s": 20.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8760859262444867, "source_start_sample": 0, "source_end_sample": 994080, "first_word_output_sample": 0, "last_word_output_sample": 986880, "samples": 994080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u00-c01.mp3" }, "de-conversation-03-podcast-block-free-u00-c02": { "id": "de-conversation-03-podcast-block-free-u00-c02", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 85603, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 20.4, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 967680, "script_start_s": 0.0, "script_end_s": 20.16, "all_asr_last_word_s": 20.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8880263269433981, "source_start_sample": 0, "source_end_sample": 974880, "first_word_output_sample": 0, "last_word_output_sample": 967680, "samples": 974880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 20.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u00-c02.mp3" }, "de-conversation-03-podcast-block-free-u00-c03": { "id": "de-conversation-03-podcast-block-free-u00-c03", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 85604, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 20.16, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 48, "coverage": 0.9795918367346939, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 960000, "script_start_s": 0.0, "script_end_s": 20.0, "all_asr_last_word_s": 20.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02040816326530612, "source_identity_reward": 0.8508180216822869, "source_start_sample": 0, "source_end_sample": 967200, "first_word_output_sample": 0, "last_word_output_sample": 960000, "samples": 967200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u00-c03.mp3" }, "de-conversation-03-podcast-block-free-u00-c04": { "id": "de-conversation-03-podcast-block-free-u00-c04", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 85605, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 19.84, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 944640, "script_start_s": 0.0, "script_end_s": 19.68, "all_asr_last_word_s": 19.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8653289990930664, "source_start_sample": 0, "source_end_sample": 951840, "first_word_output_sample": 0, "last_word_output_sample": 944640, "samples": 951840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u00-c04.mp3" }, "de-conversation-03-podcast-block-free-u00-c05": { "id": "de-conversation-03-podcast-block-free-u00-c05", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 85606, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 30.72, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1009920, "script_start_s": 0.0, "script_end_s": 21.04, "all_asr_last_word_s": 30.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.30612244897959184, "source_identity_reward": 0.6520625844331007, "source_start_sample": 0, "source_end_sample": 1017120, "first_word_output_sample": 0, "last_word_output_sample": 1009920, "samples": 1017120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 9.53, "added_silence_s": 0.0, "duration_s": 21.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u00-c05.mp3" }, "de-conversation-03-podcast-block-free-u00-c06": { "id": "de-conversation-03-podcast-block-free-u00-c06", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 85607, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 29.6, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 998400, "script_start_s": 0.0, "script_end_s": 20.8, "all_asr_last_word_s": 28.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04081632653061224, "source_identity_reward": 0.8703187274371179, "source_start_sample": 0, "source_end_sample": 1005600, "first_word_output_sample": 0, "last_word_output_sample": 998400, "samples": 1005600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 8.65, "added_silence_s": 0.0, "duration_s": 20.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u00-c06.mp3" }, "de-conversation-03-podcast-block-free-u00-c07": { "id": "de-conversation-03-podcast-block-free-u00-c07", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 85608, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können. Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt? Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf. Das wäre toll, aber gerade bei langen Gesprächen können Stimmen allmählich abdriften.", "source_duration_s": 19.76, "source_frame_cap": false, "alignment": { "expected_tokens": 49, "matched_tokens": 49, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 940800, "script_start_s": 0.0, "script_end_s": 19.6, "all_asr_last_word_s": 19.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8744623365108926, "source_start_sample": 0, "source_end_sample": 948000, "first_word_output_sample": 0, "last_word_output_sample": 940800, "samples": 948000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u00-c07.mp3" }, "de-conversation-02-podcast-block-timed-u01-c00": { "id": "de-conversation-02-podcast-block-timed-u01-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 84617, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.84, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 62, "coverage": 0.9253731343283582, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07462686567164178, "source_identity_reward": 0.653425012782462 }, "de-conversation-02-podcast-block-timed-u01-c01": { "id": "de-conversation-02-podcast-block-timed-u01-c01", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 84618, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.84, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1140480, "script_start_s": 0.0, "script_end_s": 23.76, "all_asr_last_word_s": 23.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.7985566519563286, "source_start_sample": 0, "source_end_sample": 1144320, "first_word_output_sample": 0, "last_word_output_sample": 1140480, "samples": 1147680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-timed-u01-c01.mp3" }, "de-conversation-02-podcast-block-timed-u01-c02": { "id": "de-conversation-02-podcast-block-timed-u01-c02", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 84619, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.84, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 64, "coverage": 0.9552238805970149, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1140480, "script_start_s": 0.0, "script_end_s": 23.76, "all_asr_last_word_s": 23.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04477611940298507, "source_identity_reward": 0.6972911235774145, "source_start_sample": 0, "source_end_sample": 1144320, "first_word_output_sample": 0, "last_word_output_sample": 1140480, "samples": 1147680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-timed-u01-c02.mp3" }, "de-conversation-02-podcast-block-timed-u01-c03": { "id": "de-conversation-02-podcast-block-timed-u01-c03", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 84620, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.84, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1140480, "script_start_s": 0.0, "script_end_s": 23.76, "all_asr_last_word_s": 23.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.7091724117863011, "source_start_sample": 0, "source_end_sample": 1144320, "first_word_output_sample": 0, "last_word_output_sample": 1140480, "samples": 1147680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-timed-u01-c03.mp3" }, "de-conversation-02-podcast-block-timed-u01-c04": { "id": "de-conversation-02-podcast-block-timed-u01-c04", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 84621, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.84, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 63, "coverage": 0.9402985074626866, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05970149253731343, "source_identity_reward": 0.6633838579625361 }, "de-conversation-02-podcast-block-timed-u01-c05": { "id": "de-conversation-02-podcast-block-timed-u01-c05", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 84622, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.92, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1144320, "script_start_s": 0.0, "script_end_s": 23.84, "all_asr_last_word_s": 23.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.7136200359008985, "source_start_sample": 0, "source_end_sample": 1148160, "first_word_output_sample": 0, "last_word_output_sample": 1144320, "samples": 1151520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-timed-u01-c05.mp3" }, "de-conversation-02-podcast-block-timed-u01-c06": { "id": "de-conversation-02-podcast-block-timed-u01-c06", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 84623, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.84, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 63, "coverage": 0.9402985074626866, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05970149253731343, "source_identity_reward": 0.7683772424896642 }, "de-conversation-02-podcast-block-timed-u01-c07": { "id": "de-conversation-02-podcast-block-timed-u01-c07", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 84624, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.84, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 62, "coverage": 0.9253731343283582, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07462686567164178, "source_identity_reward": 0.6571287660991182 }, "de-conversation-04-podcast-segment-timed-u11-c00": { "id": "de-conversation-04-podcast-segment-timed-u11-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 222720, "script_start_s": 0.24, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3888489602199583, "source_start_sample": 7680, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-timed-u11-c00.mp3" }, "de-conversation-05-podcast-segment-free-u12-c00": { "id": "de-conversation-05-podcast-segment-free-u12-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87293, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen.", "source_duration_s": 19.2, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 311040, "script_start_s": 0.16, "script_end_s": 6.48, "all_asr_last_word_s": 19.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u12-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.9333333333333333, "source_identity_reward": 0.3524964479906469, "source_start_sample": 3840, "source_end_sample": 318240, "first_word_output_sample": 3840, "last_word_output_sample": 307200, "samples": 314400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 12.57, "added_silence_s": 0.0, "duration_s": 6.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u12-c00.mp3" }, "de-conversation-05-podcast-block-timed-u03-c00": { "id": "de-conversation-05-podcast-block-timed-u03-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 87649, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.96, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1190400, "script_start_s": 0.0, "script_end_s": 24.8, "all_asr_last_word_s": 24.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04285714285714286, "source_identity_reward": 0.8268062712378447, "source_start_sample": 0, "source_end_sample": 1197600, "first_word_output_sample": 0, "last_word_output_sample": 1190400, "samples": 1197600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 24.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u03-c00.mp3" }, "de-conversation-05-podcast-block-timed-u03-c01": { "id": "de-conversation-05-podcast-block-timed-u03-c01", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 87650, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.96, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 68, "coverage": 0.9714285714285714, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u03-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05714285714285714, "source_identity_reward": 0.7915437515717011 }, "de-conversation-05-podcast-block-timed-u03-c02": { "id": "de-conversation-05-podcast-block-timed-u03-c02", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 87651, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.96, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1194240, "script_start_s": 0.0, "script_end_s": 24.88, "all_asr_last_word_s": 24.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u03-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04285714285714286, "source_identity_reward": 0.7749713701642106, "source_start_sample": 0, "source_end_sample": 1198080, "first_word_output_sample": 0, "last_word_output_sample": 1194240, "samples": 1201440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 25.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u03-c02.mp3" }, "de-conversation-05-podcast-block-timed-u03-c03": { "id": "de-conversation-05-podcast-block-timed-u03-c03", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 87652, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.96, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 68, "coverage": 0.9714285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1194240, "script_start_s": 0.0, "script_end_s": 24.88, "all_asr_last_word_s": 24.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u03-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05714285714285714, "source_identity_reward": 0.8089141007795031, "source_start_sample": 0, "source_end_sample": 1198080, "first_word_output_sample": 0, "last_word_output_sample": 1194240, "samples": 1201440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 25.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u03-c03.mp3" }, "de-conversation-05-podcast-block-timed-u03-c04": { "id": "de-conversation-05-podcast-block-timed-u03-c04", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 87653, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.88, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1190400, "script_start_s": 0.0, "script_end_s": 24.8, "all_asr_last_word_s": 24.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u03-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04285714285714286, "source_identity_reward": 0.8145494235751685, "source_start_sample": 0, "source_end_sample": 1194240, "first_word_output_sample": 0, "last_word_output_sample": 1190400, "samples": 1197600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 24.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u03-c04.mp3" }, "de-conversation-05-podcast-block-timed-u03-c05": { "id": "de-conversation-05-podcast-block-timed-u03-c05", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 87654, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 24.96, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1194240, "script_start_s": 0.0, "script_end_s": 24.88, "all_asr_last_word_s": 24.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u03-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04285714285714286, "source_identity_reward": 0.7849238353752828, "source_start_sample": 0, "source_end_sample": 1198080, "first_word_output_sample": 0, "last_word_output_sample": 1194240, "samples": 1201440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 25.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u03-c05.mp3" }, "de-conversation-05-podcast-block-timed-u03-c06": { "id": "de-conversation-05-podcast-block-timed-u03-c06", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 87655, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 25.04, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 69, "coverage": 0.9857142857142858, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1198080, "script_start_s": 0.0, "script_end_s": 24.96, "all_asr_last_word_s": 24.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u03-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04285714285714286, "source_identity_reward": 0.8127371100142147, "source_start_sample": 0, "source_end_sample": 1201920, "first_word_output_sample": 0, "last_word_output_sample": 1198080, "samples": 1205280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 25.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u03-c06.mp3" }, "de-conversation-05-podcast-block-timed-u03-c07": { "id": "de-conversation-05-podcast-block-timed-u03-c07", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 87656, "plain": "Stell dir eine kleine Wissenschaftssendung vor, bei der ehrliche Neugier wichtiger ist als große Versprechen. Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden. Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein. Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 25.04, "source_frame_cap": false, "alignment": { "expected_tokens": 70, "matched_tokens": 68, "coverage": 0.9714285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1194240, "script_start_s": 0.0, "script_end_s": 24.88, "all_asr_last_word_s": 24.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-timed-u03-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05714285714285714, "source_identity_reward": 0.7725570227808182, "source_start_sample": 0, "source_end_sample": 1201440, "first_word_output_sample": 0, "last_word_output_sample": 1194240, "samples": 1201440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 25.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-timed-u03-c07.mp3" }, "en-conversation-01-podcast-segment-timed-u11-c00": { "id": "en-conversation-01-podcast-segment-timed-u11-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 238080, "script_start_s": 0.24, "script_end_s": 4.96, "all_asr_last_word_s": 4.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06666666666666667, "source_identity_reward": 0.06418696280288497, "source_start_sample": 7680, "source_end_sample": 241920, "first_word_output_sample": 3840, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-timed-u11-c00.mp3" }, "de-conversation-04-podcast-segment-timed-u01-c00": { "id": "de-conversation-04-podcast-segment-timed-u01-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86117, "plain": "Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde?", "source_duration_s": 4.0, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 10, "coverage": 0.9090909090909091, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 188160, "script_start_s": 0.08, "script_end_s": 3.92, "all_asr_last_word_s": 3.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09090909090909091, "source_identity_reward": 0.3166398715556572, "source_start_sample": 0, "source_end_sample": 192000, "first_word_output_sample": 3840, "last_word_output_sample": 188160, "samples": 195360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-timed-u01-c00.mp3" }, "en-conversation-05-podcast-segment-timed-u12-c00": { "id": "en-conversation-05-podcast-segment-timed-u12-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77293, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 238080, "script_start_s": 0.16, "script_end_s": 4.96, "all_asr_last_word_s": 4.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u12-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06666666666666667, "source_identity_reward": 0.38757654141333153, "source_start_sample": 3840, "source_end_sample": 241920, "first_word_output_sample": 3840, "last_word_output_sample": 234240, "samples": 241440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-timed-u12-c00.mp3" }, "de-conversation-04-podcast-segment-free-u05-c00": { "id": "de-conversation-04-podcast-segment-free-u05-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86181, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen.", "source_duration_s": 8.8, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 268800, "script_start_s": 0.0, "script_end_s": 5.6, "all_asr_last_word_s": 5.6000000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6671356824727556, "source_start_sample": 0, "source_end_sample": 276000, "first_word_output_sample": 0, "last_word_output_sample": 268800, "samples": 276000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 3.05, "added_silence_s": 0.0, "duration_s": 5.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-free-u05-c00.mp3" }, "de-conversation-01-podcast-segment-free-u06-c00": { "id": "de-conversation-01-podcast-segment-free-u06-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83197, "plain": "Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 303360, "script_start_s": 0.0, "script_end_s": 6.32, "all_asr_last_word_s": 6.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6979960067361779, "source_start_sample": 0, "source_end_sample": 310560, "first_word_output_sample": 0, "last_word_output_sample": 303360, "samples": 310560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.53, "added_silence_s": 0.0, "duration_s": 6.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-free-u06-c00.mp3" }, "de-conversation-01-podcast-segment-free-u10-c00": { "id": "de-conversation-01-podcast-segment-free-u10-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 7.68, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 15, "coverage": 0.9375, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0625, "source_identity_reward": 0.8620243518316184 }, "en-conversation-05-podcast-block-timed-u00-c00": { "id": "en-conversation-05-podcast-block-timed-u00-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 77601, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.0, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 53, "coverage": 0.9636363636363636, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 860160, "script_start_s": 0.08, "script_end_s": 17.92, "all_asr_last_word_s": 17.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03636363636363636, "source_identity_reward": 0.5278148853317948, "source_start_sample": 0, "source_end_sample": 864000, "first_word_output_sample": 3840, "last_word_output_sample": 860160, "samples": 867360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u00-c00.mp3" }, "en-conversation-05-podcast-block-timed-u00-c01": { "id": "en-conversation-05-podcast-block-timed-u00-c01", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 77602, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.0, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 53, "coverage": 0.9636363636363636, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 860160, "script_start_s": 0.08, "script_end_s": 17.92, "all_asr_last_word_s": 17.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03636363636363636, "source_identity_reward": 0.6945034887771617, "source_start_sample": 0, "source_end_sample": 864000, "first_word_output_sample": 3840, "last_word_output_sample": 860160, "samples": 867360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u00-c01.mp3" }, "en-conversation-05-podcast-block-timed-u00-c02": { "id": "en-conversation-05-podcast-block-timed-u00-c02", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 77603, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.0, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 53, "coverage": 0.9636363636363636, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 860160, "script_start_s": 0.08, "script_end_s": 17.92, "all_asr_last_word_s": 17.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03636363636363636, "source_identity_reward": 0.5072060701969545, "source_start_sample": 0, "source_end_sample": 864000, "first_word_output_sample": 3840, "last_word_output_sample": 860160, "samples": 867360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u00-c02.mp3" }, "en-conversation-05-podcast-block-timed-u00-c03": { "id": "en-conversation-05-podcast-block-timed-u00-c03", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 77604, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.0, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 53, "coverage": 0.9636363636363636, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 860160, "script_start_s": 0.0, "script_end_s": 17.92, "all_asr_last_word_s": 17.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03636363636363636, "source_identity_reward": 0.5217613916965359, "source_start_sample": 0, "source_end_sample": 864000, "first_word_output_sample": 0, "last_word_output_sample": 860160, "samples": 867360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u00-c03.mp3" }, "en-conversation-05-podcast-block-timed-u00-c04": { "id": "en-conversation-05-podcast-block-timed-u00-c04", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 77605, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.0, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 53, "coverage": 0.9636363636363636, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03636363636363636, "source_identity_reward": 0.505338149590936 }, "en-conversation-05-podcast-block-timed-u00-c05": { "id": "en-conversation-05-podcast-block-timed-u00-c05", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 77606, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.0, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 53, "coverage": 0.9636363636363636, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 860160, "script_start_s": 0.0, "script_end_s": 17.92, "all_asr_last_word_s": 17.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03636363636363636, "source_identity_reward": 0.5616179915724133, "source_start_sample": 0, "source_end_sample": 864000, "first_word_output_sample": 0, "last_word_output_sample": 860160, "samples": 867360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u00-c05.mp3" }, "en-conversation-05-podcast-block-timed-u00-c06": { "id": "en-conversation-05-podcast-block-timed-u00-c06", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 77607, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.0, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 860160, "script_start_s": 0.16, "script_end_s": 17.92, "all_asr_last_word_s": 17.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.6025492328046058, "source_start_sample": 3840, "source_end_sample": 864000, "first_word_output_sample": 3840, "last_word_output_sample": 856320, "samples": 863520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u00-c06.mp3" }, "en-conversation-05-podcast-block-timed-u00-c07": { "id": "en-conversation-05-podcast-block-timed-u00-c07", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 77608, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 18.0, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 860160, "script_start_s": 0.0, "script_end_s": 17.92, "all_asr_last_word_s": 17.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4560333374980206, "source_start_sample": 0, "source_end_sample": 864000, "first_word_output_sample": 0, "last_word_output_sample": 860160, "samples": 867360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u00-c07.mp3" }, "en-conversation-05-podcast-segment-timed-u04-c00": { "id": "en-conversation-05-podcast-segment-timed-u04-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77165, "plain": "Then try three, clearly marked as exploratory because training focused on two speakers.", "source_duration_s": 4.56, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 215040, "script_start_s": 0.0, "script_end_s": 4.48, "all_asr_last_word_s": 4.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1890138776227746, "source_start_sample": 0, "source_end_sample": 218880, "first_word_output_sample": 0, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-timed-u04-c00.mp3" }, "en-conversation-02-podcast-block-free-u02-c00": { "id": "en-conversation-02-podcast-block-free-u02-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 74633, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.64, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 483840, "script_start_s": 0.0, "script_end_s": 10.08, "all_asr_last_word_s": 18.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.6206896551724138, "source_identity_reward": 0.29411791335480225, "source_start_sample": 0, "source_end_sample": 491040, "first_word_output_sample": 0, "last_word_output_sample": 483840, "samples": 491040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 8.41, "added_silence_s": 0.0, "duration_s": 10.23, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u02-c00.mp3" }, "en-conversation-02-podcast-block-free-u02-c01": { "id": "en-conversation-02-podcast-block-free-u02-c01", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 74634, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 12.0, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 495360, "script_start_s": 0.0, "script_end_s": 10.32, "all_asr_last_word_s": 10.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.47911915344062855, "source_start_sample": 0, "source_end_sample": 502560, "first_word_output_sample": 0, "last_word_output_sample": 495360, "samples": 502560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.53, "added_silence_s": 0.0, "duration_s": 10.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u02-c01.mp3" }, "en-conversation-02-podcast-block-free-u02-c02": { "id": "en-conversation-02-podcast-block-free-u02-c02", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 74635, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 15.36, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 476160, "script_start_s": 0.16, "script_end_s": 9.92, "all_asr_last_word_s": 14.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.41379310344827586, "source_identity_reward": 0.38698090834311033, "source_start_sample": 3840, "source_end_sample": 483360, "first_word_output_sample": 3840, "last_word_output_sample": 472320, "samples": 479520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 5.29, "added_silence_s": 0.0, "duration_s": 9.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u02-c02.mp3" }, "en-conversation-02-podcast-block-free-u02-c03": { "id": "en-conversation-02-podcast-block-free-u02-c03", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 74636, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 14.8, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 449280, "script_start_s": 0.0, "script_end_s": 9.36, "all_asr_last_word_s": 14.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5172413793103449, "source_identity_reward": 0.4264341299510857, "source_start_sample": 0, "source_end_sample": 456480, "first_word_output_sample": 0, "last_word_output_sample": 449280, "samples": 456480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 5.29, "added_silence_s": 0.0, "duration_s": 9.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u02-c03.mp3" }, "en-conversation-02-podcast-block-free-u02-c04": { "id": "en-conversation-02-podcast-block-free-u02-c04", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 74637, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 456960, "script_start_s": 0.0, "script_end_s": 9.52, "all_asr_last_word_s": 59.519999999999996 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 2.1379310344827585, "source_identity_reward": 0.17542679219248028, "source_start_sample": 0, "source_end_sample": 464160, "first_word_output_sample": 0, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 50.33, "added_silence_s": 0.0, "duration_s": 9.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u02-c04.mp3" }, "en-conversation-02-podcast-block-free-u02-c05": { "id": "en-conversation-02-podcast-block-free-u02-c05", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 74638, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.2, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 433920, "script_start_s": 0.16, "script_end_s": 9.04, "all_asr_last_word_s": 9.040000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.49589186993231005, "source_start_sample": 3840, "source_end_sample": 441120, "first_word_output_sample": 3840, "last_word_output_sample": 430080, "samples": 437280, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u02-c05.mp3" }, "en-conversation-02-podcast-block-free-u02-c06": { "id": "en-conversation-02-podcast-block-free-u02-c06", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 74639, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 11.36, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 472320, "script_start_s": 0.16, "script_end_s": 9.84, "all_asr_last_word_s": 9.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6689207530253412, "source_start_sample": 3840, "source_end_sample": 479520, "first_word_output_sample": 3840, "last_word_output_sample": 468480, "samples": 475680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.37, "added_silence_s": 0.0, "duration_s": 9.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u02-c06.mp3" }, "en-conversation-02-podcast-block-free-u02-c07": { "id": "en-conversation-02-podcast-block-free-u02-c07", "case": "en-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 74640, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 466560, "script_start_s": 0.0, "script_end_s": 9.72, "all_asr_last_word_s": 9.719999999999999 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6584605192696793, "source_start_sample": 0, "source_end_sample": 473760, "first_word_output_sample": 0, "last_word_output_sample": 466560, "samples": 473760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 50.13, "added_silence_s": 0.0, "duration_s": 9.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-block-free-u02-c07.mp3" }, "en-conversation-01-podcast-block-timed-u02-c00": { "id": "en-conversation-01-podcast-block-timed-u02-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 73633, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 875520, "script_start_s": 0.0, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4383768956977645, "source_start_sample": 0, "source_end_sample": 879360, "first_word_output_sample": 0, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u02-c00.mp3" }, "en-conversation-01-podcast-block-timed-u02-c01": { "id": "en-conversation-01-podcast-block-timed-u02-c01", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 73634, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 875520, "script_start_s": 0.16, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4584330388633059, "source_start_sample": 3840, "source_end_sample": 879360, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u02-c01.mp3" }, "en-conversation-01-podcast-block-timed-u02-c02": { "id": "en-conversation-01-podcast-block-timed-u02-c02", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 73635, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 871680, "script_start_s": 0.16, "script_end_s": 18.16, "all_asr_last_word_s": 18.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.5421950998750827, "source_start_sample": 3840, "source_end_sample": 878880, "first_word_output_sample": 3840, "last_word_output_sample": 867840, "samples": 875040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 18.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u02-c02.mp3" }, "en-conversation-01-podcast-block-timed-u02-c03": { "id": "en-conversation-01-podcast-block-timed-u02-c03", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 73636, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 875520, "script_start_s": 0.0, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4934639721810574, "source_start_sample": 0, "source_end_sample": 879360, "first_word_output_sample": 0, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u02-c03.mp3" }, "en-conversation-01-podcast-block-timed-u02-c04": { "id": "en-conversation-01-podcast-block-timed-u02-c04", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 73637, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 879360, "script_start_s": 0.0, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.5115497064410995, "source_start_sample": 0, "source_end_sample": 883200, "first_word_output_sample": 0, "last_word_output_sample": 879360, "samples": 886560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u02-c04.mp3" }, "en-conversation-01-podcast-block-timed-u02-c05": { "id": "en-conversation-01-podcast-block-timed-u02-c05", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 73638, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 875520, "script_start_s": 0.16, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4307986211372602, "source_start_sample": 3840, "source_end_sample": 879360, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u02-c05.mp3" }, "en-conversation-01-podcast-block-timed-u02-c06": { "id": "en-conversation-01-podcast-block-timed-u02-c06", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 73639, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 875520, "script_start_s": 0.0, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3865208427213528, "source_start_sample": 0, "source_end_sample": 879360, "first_word_output_sample": 0, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u02-c06.mp3" }, "en-conversation-01-podcast-block-timed-u02-c07": { "id": "en-conversation-01-podcast-block-timed-u02-c07", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 73640, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 875520, "script_start_s": 0.16, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.4087676340797771, "source_start_sample": 3840, "source_end_sample": 879360, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u02-c07.mp3" }, "de-conversation-04-podcast-segment-free-u07-c00": { "id": "de-conversation-04-podcast-segment-free-u07-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86213, "plain": "Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine.", "source_duration_s": 8.88, "source_frame_cap": false, "alignment": { "expected_tokens": 10, "matched_tokens": 10, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 264960, "script_start_s": 0.24, "script_end_s": 5.52, "all_asr_last_word_s": 5.5200000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5586212451316671, "source_start_sample": 7680, "source_end_sample": 272160, "first_word_output_sample": 3840, "last_word_output_sample": 257280, "samples": 264480, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 3.21, "added_silence_s": 0.0, "duration_s": 5.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-free-u07-c00.mp3" }, "en-conversation-01-podcast-segment-timed-u08-c00": { "id": "en-conversation-01-podcast-segment-timed-u08-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 222720, "script_start_s": 0.16, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.23515354949457695, "source_start_sample": 3840, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-timed-u08-c00.mp3" }, "en-conversation-03-podcast-block-timed-u01-c00": { "id": "en-conversation-03-podcast-block-timed-u01-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 75617, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.88, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 802560, "script_start_s": 0.0, "script_end_s": 16.72, "all_asr_last_word_s": 16.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5000071569149818, "source_start_sample": 0, "source_end_sample": 809760, "first_word_output_sample": 0, "last_word_output_sample": 802560, "samples": 809760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u01-c00.mp3" }, "en-conversation-03-podcast-block-timed-u01-c01": { "id": "en-conversation-03-podcast-block-timed-u01-c01", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 75618, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 806400, "script_start_s": 0.0, "script_end_s": 16.8, "all_asr_last_word_s": 16.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.45069895921340297, "source_start_sample": 0, "source_end_sample": 813600, "first_word_output_sample": 0, "last_word_output_sample": 806400, "samples": 813600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u01-c01.mp3" }, "en-conversation-03-podcast-block-timed-u01-c02": { "id": "en-conversation-03-podcast-block-timed-u01-c02", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 75619, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 50, "coverage": 0.9803921568627451, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 810240, "script_start_s": 0.0, "script_end_s": 16.88, "all_asr_last_word_s": 16.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0196078431372549, "source_identity_reward": 0.5070673538110282, "source_start_sample": 0, "source_end_sample": 814080, "first_word_output_sample": 0, "last_word_output_sample": 810240, "samples": 817440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u01-c02.mp3" }, "en-conversation-03-podcast-block-timed-u01-c03": { "id": "en-conversation-03-podcast-block-timed-u01-c03", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 75620, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.88, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 48, "coverage": 0.9411764705882353, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 806400, "script_start_s": 0.0, "script_end_s": 16.8, "all_asr_last_word_s": 16.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0784313725490196, "source_identity_reward": 0.45009288545027026, "source_start_sample": 0, "source_end_sample": 810240, "first_word_output_sample": 0, "last_word_output_sample": 806400, "samples": 813600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u01-c03.mp3" }, "en-conversation-03-podcast-block-timed-u01-c04": { "id": "en-conversation-03-podcast-block-timed-u01-c04", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 75621, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 810240, "script_start_s": 0.0, "script_end_s": 16.88, "all_asr_last_word_s": 16.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5752354757148249, "source_start_sample": 0, "source_end_sample": 814080, "first_word_output_sample": 0, "last_word_output_sample": 810240, "samples": 817440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u01-c04.mp3" }, "en-conversation-03-podcast-block-timed-u01-c05": { "id": "en-conversation-03-podcast-block-timed-u01-c05", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 75622, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 806400, "script_start_s": 0.0, "script_end_s": 16.8, "all_asr_last_word_s": 16.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5521524862379532, "source_start_sample": 0, "source_end_sample": 813600, "first_word_output_sample": 0, "last_word_output_sample": 806400, "samples": 813600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u01-c05.mp3" }, "en-conversation-03-podcast-block-timed-u01-c06": { "id": "en-conversation-03-podcast-block-timed-u01-c06", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 75623, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 49, "coverage": 0.9607843137254902, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 806400, "script_start_s": 0.16, "script_end_s": 16.8, "all_asr_last_word_s": 16.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0784313725490196, "source_identity_reward": 0.5115574894222643, "source_start_sample": 3840, "source_end_sample": 813600, "first_word_output_sample": 3840, "last_word_output_sample": 802560, "samples": 809760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u01-c06.mp3" }, "en-conversation-03-podcast-block-timed-u01-c07": { "id": "en-conversation-03-podcast-block-timed-u01-c07", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 75624, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 810240, "script_start_s": 0.16, "script_end_s": 16.88, "all_asr_last_word_s": 16.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.41704176446320945, "source_start_sample": 3840, "source_end_sample": 814080, "first_word_output_sample": 3840, "last_word_output_sample": 806400, "samples": 813600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u01-c07.mp3" }, "en-conversation-03-podcast-segment-timed-u09-c00": { "id": "en-conversation-03-podcast-segment-timed-u09-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 192000, "script_start_s": 0.16, "script_end_s": 4.0, "all_asr_last_word_s": 4.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2580842605011703, "source_start_sample": 3840, "source_end_sample": 195840, "first_word_output_sample": 3840, "last_word_output_sample": 188160, "samples": 195360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-timed-u09-c00.mp3" }, "en-conversation-01-podcast-segment-free-u01-c00": { "id": "en-conversation-01-podcast-segment-free-u01-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73117, "plain": "You sound suspiciously excited. Did a speech model actually surprise you?", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 249600, "script_start_s": 0.16, "script_end_s": 5.2, "all_asr_last_word_s": 5.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3533014899366216, "source_start_sample": 3840, "source_end_sample": 256800, "first_word_output_sample": 3840, "last_word_output_sample": 245760, "samples": 252960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 24.65, "added_silence_s": 0.0, "duration_s": 5.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-free-u01-c00.mp3" }, "de-conversation-04-podcast-segment-timed-u10-c00": { "id": "de-conversation-04-podcast-segment-timed-u10-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 5.76, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 272640, "script_start_s": 0.0, "script_end_s": 5.68, "all_asr_last_word_s": 5.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6535980163120418, "source_start_sample": 0, "source_end_sample": 276480, "first_word_output_sample": 0, "last_word_output_sample": 272640, "samples": 279840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-timed-u10-c00.mp3" }, "de-conversation-03-podcast-segment-timed-u07-c00": { "id": "de-conversation-03-podcast-segment-timed-u07-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85213, "plain": "Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 192000, "script_start_s": 0.08, "script_end_s": 4.0, "all_asr_last_word_s": 4.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4640739349054904, "source_start_sample": 0, "source_end_sample": 195840, "first_word_output_sample": 3840, "last_word_output_sample": 192000, "samples": 199200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-timed-u07-c00.mp3" }, "en-conversation-03-podcast-segment-free-u01-c00": { "id": "en-conversation-03-podcast-segment-free-u01-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75117, "plain": "Then keeping those voices distinct while the conversation wanders somewhere unexpected?", "source_duration_s": 6.32, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 288000, "script_start_s": 0.16, "script_end_s": 6.0, "all_asr_last_word_s": 6.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2967026136515839, "source_start_sample": 3840, "source_end_sample": 295200, "first_word_output_sample": 3840, "last_word_output_sample": 284160, "samples": 291360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.17, "added_silence_s": 0.0, "duration_s": 6.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-free-u01-c00.mp3" }, "de-conversation-01-podcast-block-timed-u01-c00": { "id": "de-conversation-01-podcast-block-timed-u01-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 83617, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 17.92, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 46, "coverage": 0.92, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08, "source_identity_reward": 0.7565954166690536 }, "de-conversation-01-podcast-block-timed-u01-c01": { "id": "de-conversation-01-podcast-block-timed-u01-c01", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 83618, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 17.84, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 48, "coverage": 0.96, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 852480, "script_start_s": 0.0, "script_end_s": 17.76, "all_asr_last_word_s": 17.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04, "source_identity_reward": 0.7362885804337445, "source_start_sample": 0, "source_end_sample": 856320, "first_word_output_sample": 0, "last_word_output_sample": 852480, "samples": 859680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u01-c01.mp3" }, "de-conversation-01-podcast-block-timed-u01-c02": { "id": "de-conversation-01-podcast-block-timed-u01-c02", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 83619, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 17.84, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 852480, "script_start_s": 0.0, "script_end_s": 17.76, "all_asr_last_word_s": 17.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.7264846158707823, "source_start_sample": 0, "source_end_sample": 856320, "first_word_output_sample": 0, "last_word_output_sample": 852480, "samples": 859680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u01-c02.mp3" }, "de-conversation-01-podcast-block-timed-u01-c03": { "id": "de-conversation-01-podcast-block-timed-u01-c03", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 83620, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 17.84, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 848640, "script_start_s": 0.0, "script_end_s": 17.68, "all_asr_last_word_s": 17.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.7808718587866795, "source_start_sample": 0, "source_end_sample": 855840, "first_word_output_sample": 0, "last_word_output_sample": 848640, "samples": 855840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u01-c03.mp3" }, "de-conversation-01-podcast-block-timed-u01-c04": { "id": "de-conversation-01-podcast-block-timed-u01-c04", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 83621, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 17.92, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 48, "coverage": 0.96, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.04, "source_identity_reward": 0.7668810221500205 }, "de-conversation-01-podcast-block-timed-u01-c05": { "id": "de-conversation-01-podcast-block-timed-u01-c05", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 83622, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 17.92, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 856320, "script_start_s": 0.0, "script_end_s": 17.84, "all_asr_last_word_s": 17.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.7382241255705474, "source_start_sample": 0, "source_end_sample": 860160, "first_word_output_sample": 0, "last_word_output_sample": 856320, "samples": 863520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u01-c05.mp3" }, "de-conversation-01-podcast-block-timed-u01-c06": { "id": "de-conversation-01-podcast-block-timed-u01-c06", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 83623, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 17.92, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 856320, "script_start_s": 0.0, "script_end_s": 17.84, "all_asr_last_word_s": 17.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.7341822576074115, "source_start_sample": 0, "source_end_sample": 860160, "first_word_output_sample": 0, "last_word_output_sample": 856320, "samples": 863520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u01-c06.mp3" }, "de-conversation-01-podcast-block-timed-u01-c07": { "id": "de-conversation-01-podcast-block-timed-u01-c07", "case": "de-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 83624, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt. Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir. Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben. Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 17.92, "source_frame_cap": false, "alignment": { "expected_tokens": 50, "matched_tokens": 49, "coverage": 0.98, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 852480, "script_start_s": 0.0, "script_end_s": 17.76, "all_asr_last_word_s": 17.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02, "source_identity_reward": 0.7661333428262493, "source_start_sample": 0, "source_end_sample": 859680, "first_word_output_sample": 0, "last_word_output_sample": 852480, "samples": 859680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-block-timed-u01-c07.mp3" }, "de-conversation-04-podcast-segment-timed-u05-c00": { "id": "de-conversation-04-podcast-segment-timed-u05-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86181, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen.", "source_duration_s": 5.36, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 249600, "script_start_s": 0.0, "script_end_s": 5.2, "all_asr_last_word_s": 5.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.43855279884267173, "source_start_sample": 0, "source_end_sample": 256800, "first_word_output_sample": 0, "last_word_output_sample": 249600, "samples": 256800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-timed-u05-c00.mp3" }, "en-conversation-05-podcast-segment-free-u10-c00": { "id": "en-conversation-05-podcast-segment-free-u10-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 257280, "script_start_s": 0.0, "script_end_s": 5.36, "all_asr_last_word_s": 29.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 3.2857142857142856, "source_identity_reward": 0.09328512018479199, "source_start_sample": 0, "source_end_sample": 264480, "first_word_output_sample": 0, "last_word_output_sample": 257280, "samples": 264480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.49, "added_silence_s": 0.0, "duration_s": 5.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-free-u10-c00.mp3" }, "de-conversation-05-podcast-segment-timed-u10-c00": { "id": "de-conversation-05-podcast-segment-timed-u10-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 5.76, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 272640, "script_start_s": 0.0, "script_end_s": 5.68, "all_asr_last_word_s": 5.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7301868303836319, "source_start_sample": 0, "source_end_sample": 276480, "first_word_output_sample": 0, "last_word_output_sample": 272640, "samples": 279840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-timed-u10-c00.mp3" }, "de-conversation-02-podcast-segment-free-u01-c00": { "id": "de-conversation-02-podcast-segment-free-u01-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84117, "plain": "Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 253440, "script_start_s": 0.16, "script_end_s": 5.28, "all_asr_last_word_s": 18.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 2.272727272727273, "source_identity_reward": 0.10640674229859132, "source_start_sample": 3840, "source_end_sample": 260640, "first_word_output_sample": 3840, "last_word_output_sample": 249600, "samples": 256800, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 24.57, "added_silence_s": 0.0, "duration_s": 5.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-free-u01-c00.mp3" }, "de-conversation-01-podcast-segment-timed-u02-c00": { "id": "de-conversation-01-podcast-segment-timed-u02-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83133, "plain": "Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz.", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 10, "coverage": 0.8333333333333334, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.16666666666666666, "source_identity_reward": 0.4614124070815183 }, "en-conversation-03-podcast-segment-free-u00-c00": { "id": "en-conversation-03-podcast-segment-free-u00-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75101, "plain": "Here is the part that interests me: using a short recording to establish each voice.", "source_duration_s": 7.36, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 253440, "script_start_s": 0.0, "script_end_s": 5.28, "all_asr_last_word_s": 5.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06666666666666667, "source_identity_reward": 0.2470907905480983, "source_start_sample": 0, "source_end_sample": 260640, "first_word_output_sample": 0, "last_word_output_sample": 253440, "samples": 260640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.93, "added_silence_s": 0.0, "duration_s": 5.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-free-u00-c00.mp3" }, "de-conversation-02-podcast-segment-timed-u04-c00": { "id": "de-conversation-02-podcast-segment-timed-u04-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84165, "plain": "Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 5.44, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 257280, "script_start_s": 0.16, "script_end_s": 5.36, "all_asr_last_word_s": 5.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6696736839052485, "source_start_sample": 3840, "source_end_sample": 261120, "first_word_output_sample": 3840, "last_word_output_sample": 253440, "samples": 260640, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-timed-u04-c00.mp3" }, "de-conversation-01-podcast-segment-timed-u04-c00": { "id": "de-conversation-01-podcast-segment-timed-u04-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83165, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt.", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 11, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 207360, "script_start_s": 0.0, "script_end_s": 4.32, "all_asr_last_word_s": 4.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.6589982666194066, "source_start_sample": 0, "source_end_sample": 211200, "first_word_output_sample": 0, "last_word_output_sample": 207360, "samples": 214560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-timed-u04-c00.mp3" }, "en-conversation-05-podcast-segment-free-u01-c00": { "id": "en-conversation-05-podcast-segment-free-u01-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77117, "plain": "Same model family, different voices, and a different emotional rhythm in each language?", "source_duration_s": 8.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 253440, "script_start_s": 0.24, "script_end_s": 5.28, "all_asr_last_word_s": 5.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3836885698269973, "source_start_sample": 7680, "source_end_sample": 260640, "first_word_output_sample": 3840, "last_word_output_sample": 245760, "samples": 252960, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 3.29, "added_silence_s": 0.0, "duration_s": 5.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-free-u01-c00.mp3" }, "de-conversation-02-base17-block-free-u02-c00": { "id": "de-conversation-02-base17-block-free-u02-c00", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 84633, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 22.56, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 572160, "script_start_s": 0.0, "script_end_s": 11.92, "all_asr_last_word_s": 11.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.22644052434129483, "source_start_sample": 0, "source_end_sample": 579360, "first_word_output_sample": 0, "last_word_output_sample": 572160, "samples": 579360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 10.49, "added_silence_s": 0.0, "duration_s": 12.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u02-c00.mp3" }, "de-conversation-02-base17-block-free-u02-c01": { "id": "de-conversation-02-base17-block-free-u02-c01", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 84634, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 11.84, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 560640, "script_start_s": 0.08, "script_end_s": 11.68, "all_asr_last_word_s": 11.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3580623154056216, "source_start_sample": 0, "source_end_sample": 567840, "first_word_output_sample": 3840, "last_word_output_sample": 560640, "samples": 567840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 11.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u02-c01.mp3" }, "de-conversation-02-base17-block-free-u02-c02": { "id": "de-conversation-02-base17-block-free-u02-c02", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 84635, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 27.28, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 549120, "script_start_s": 0.0, "script_end_s": 11.44, "all_asr_last_word_s": 23.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3448275862068966, "source_identity_reward": 0.16697699367270774, "source_start_sample": 0, "source_end_sample": 556320, "first_word_output_sample": 0, "last_word_output_sample": 549120, "samples": 556320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 15.69, "added_silence_s": 0.0, "duration_s": 11.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u02-c02.mp3" }, "de-conversation-02-base17-block-free-u02-c03": { "id": "de-conversation-02-base17-block-free-u02-c03", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 84636, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 16.0, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 576000, "script_start_s": 0.0, "script_end_s": 12.0, "all_asr_last_word_s": 12.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2443392070299502, "source_start_sample": 0, "source_end_sample": 583200, "first_word_output_sample": 0, "last_word_output_sample": 576000, "samples": 583200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 3.85, "added_silence_s": 0.0, "duration_s": 12.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u02-c03.mp3" }, "de-conversation-02-base17-block-free-u02-c04": { "id": "de-conversation-02-base17-block-free-u02-c04", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 84637, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 526080, "script_start_s": 0.0, "script_end_s": 10.96, "all_asr_last_word_s": 10.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.21896977999000364, "source_start_sample": 0, "source_end_sample": 533280, "first_word_output_sample": 0, "last_word_output_sample": 526080, "samples": 533280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 48.89, "added_silence_s": 0.0, "duration_s": 11.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u02-c04.mp3" }, "de-conversation-02-base17-block-free-u02-c05": { "id": "de-conversation-02-base17-block-free-u02-c05", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 84638, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 576000, "script_start_s": 0.24, "script_end_s": 12.0, "all_asr_last_word_s": 20.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4827586206896552, "source_identity_reward": 0.20766289168517302, "source_start_sample": 7680, "source_end_sample": 583200, "first_word_output_sample": 3840, "last_word_output_sample": 568320, "samples": 575520, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 47.85, "added_silence_s": 0.0, "duration_s": 11.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u02-c05.mp3" }, "de-conversation-02-base17-block-free-u02-c06": { "id": "de-conversation-02-base17-block-free-u02-c06", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 84639, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 568320, "script_start_s": 0.16, "script_end_s": 11.84, "all_asr_last_word_s": 51.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3793103448275862, "source_identity_reward": 0.14145074734432095, "source_start_sample": 3840, "source_end_sample": 575520, "first_word_output_sample": 3840, "last_word_output_sample": 564480, "samples": 571680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 48.01, "added_silence_s": 0.0, "duration_s": 11.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u02-c06.mp3" }, "de-conversation-02-base17-block-free-u02-c07": { "id": "de-conversation-02-base17-block-free-u02-c07", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 84640, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 14.4, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 572160, "script_start_s": 0.0, "script_end_s": 11.92, "all_asr_last_word_s": 11.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.25187477289613175, "source_start_sample": 0, "source_end_sample": 579360, "first_word_output_sample": 0, "last_word_output_sample": 572160, "samples": 579360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.33, "added_silence_s": 0.0, "duration_s": 12.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-free-u02-c07.mp3" }, "en-conversation-03-base17-segment-free-u11-c00": { "id": "en-conversation-03-base17-segment-free-u11-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 15, "matched_tokens": 14, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 272640, "script_start_s": 0.16, "script_end_s": 5.68, "all_asr_last_word_s": 5.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06666666666666667, "source_identity_reward": 0.11361288119360044, "source_start_sample": 3840, "source_end_sample": 279840, "first_word_output_sample": 3840, "last_word_output_sample": 268800, "samples": 276000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 24.17, "added_silence_s": 0.0, "duration_s": 5.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-free-u11-c00.mp3" }, "en-conversation-02-base17-segment-free-u05-c00": { "id": "en-conversation-02-base17-segment-free-u05-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74181, "plain": "I refuse to be dazzled by a cheerful description. Let me hear the voices change.", "source_duration_s": 10.24, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 483840, "script_start_s": 0.24, "script_end_s": 10.08, "all_asr_last_word_s": 10.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2276392294998275, "source_start_sample": 7680, "source_end_sample": 491040, "first_word_output_sample": 3840, "last_word_output_sample": 476160, "samples": 483360, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-free-u05-c00.mp3" }, "en-conversation-01-base17-segment-timed-u04-c00": { "id": "en-conversation-01-base17-segment-timed-u04-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73165, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances.", "source_duration_s": 4.0, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 10, "coverage": 0.8333333333333334, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 188160, "script_start_s": 0.16, "script_end_s": 3.92, "all_asr_last_word_s": 3.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.16666666666666666, "source_identity_reward": 0.3643587763904844, "source_start_sample": 3840, "source_end_sample": 192000, "first_word_output_sample": 3840, "last_word_output_sample": 184320, "samples": 191520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-timed-u04-c00.mp3" }, "en-conversation-05-base17-segment-free-u14-c00": { "id": "en-conversation-05-base17-segment-free-u14-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77325, "plain": "Exactly. The exchange should feel responsive, with the same two people returning after every change of mood.", "source_duration_s": 9.76, "source_frame_cap": false, "alignment": { "expected_tokens": 17, "matched_tokens": 17, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 307200, "script_start_s": 0.16, "script_end_s": 6.4, "all_asr_last_word_s": 6.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u14-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6021349992913552, "source_start_sample": 3840, "source_end_sample": 314400, "first_word_output_sample": 3840, "last_word_output_sample": 303360, "samples": 310560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 3.21, "added_silence_s": 0.0, "duration_s": 6.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-free-u14-c00.mp3" }, "en-conversation-05-base17-segment-free-u07-c00": { "id": "en-conversation-05-base17-segment-free-u07-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77213, "plain": "All right. I am still skeptical, but now I really want to hear the next turn.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 16, "matched_tokens": 13, "coverage": 0.8125, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.625, "source_identity_reward": 0.06882001849660829 }, "de-conversation-05-base17-block-timed-u02-c00": { "id": "de-conversation-05-base17-block-timed-u02-c00", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 87633, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 952320, "script_start_s": 0.16, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3566106255502205, "source_start_sample": 3840, "source_end_sample": 959520, "first_word_output_sample": 3840, "last_word_output_sample": 948480, "samples": 955680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u02-c00.mp3" }, "de-conversation-05-base17-block-timed-u02-c01": { "id": "de-conversation-05-base17-block-timed-u02-c01", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 87634, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 54, "coverage": 0.9642857142857143, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 956160, "script_start_s": 0.24, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.24734380279553816, "source_start_sample": 7680, "source_end_sample": 960000, "first_word_output_sample": 3840, "last_word_output_sample": 948480, "samples": 955680, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 19.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u02-c01.mp3" }, "de-conversation-05-base17-block-timed-u02-c02": { "id": "de-conversation-05-base17-block-timed-u02-c02", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 87635, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 54, "coverage": 0.9642857142857143, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 956160, "script_start_s": 0.24, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.30969872558378175, "source_start_sample": 7680, "source_end_sample": 960000, "first_word_output_sample": 3840, "last_word_output_sample": 948480, "samples": 955680, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 19.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u02-c02.mp3" }, "de-conversation-05-base17-block-timed-u02-c03": { "id": "de-conversation-05-base17-block-timed-u02-c03", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 87636, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 952320, "script_start_s": 0.16, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.47576069902342943, "source_start_sample": 3840, "source_end_sample": 959520, "first_word_output_sample": 3840, "last_word_output_sample": 948480, "samples": 955680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u02-c03.mp3" }, "de-conversation-05-base17-block-timed-u02-c04": { "id": "de-conversation-05-base17-block-timed-u02-c04", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 87637, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 952320, "script_start_s": 0.16, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.224052390132618, "source_start_sample": 3840, "source_end_sample": 959520, "first_word_output_sample": 3840, "last_word_output_sample": 948480, "samples": 955680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u02-c04.mp3" }, "de-conversation-05-base17-block-timed-u02-c05": { "id": "de-conversation-05-base17-block-timed-u02-c05", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 87638, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 54, "coverage": 0.9642857142857143, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 952320, "script_start_s": 0.16, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.2913102049972616, "source_start_sample": 3840, "source_end_sample": 959520, "first_word_output_sample": 3840, "last_word_output_sample": 948480, "samples": 955680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u02-c05.mp3" }, "de-conversation-05-base17-block-timed-u02-c06": { "id": "de-conversation-05-base17-block-timed-u02-c06", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 87639, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 956160, "script_start_s": 0.16, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3768761173900743, "source_start_sample": 3840, "source_end_sample": 960000, "first_word_output_sample": 3840, "last_word_output_sample": 952320, "samples": 959520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 19.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u02-c06.mp3" }, "de-conversation-05-base17-block-timed-u02-c07": { "id": "de-conversation-05-base17-block-timed-u02-c07", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 87640, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 952320, "script_start_s": 0.16, "script_end_s": 19.84, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2006190070530966, "source_start_sample": 3840, "source_end_sample": 959520, "first_word_output_sample": 3840, "last_word_output_sample": 948480, "samples": 955680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u02-c07.mp3" }, "en-conversation-05-base17-segment-free-u00-c00": { "id": "en-conversation-05-base17-segment-free-u00-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77101, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 472320, "script_start_s": 0.16, "script_end_s": 9.84, "all_asr_last_word_s": 13.280000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.15384615384615385, "source_identity_reward": 0.322529220585573, "source_start_sample": 3840, "source_end_sample": 479520, "first_word_output_sample": 3840, "last_word_output_sample": 468480, "samples": 475680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 20.01, "added_silence_s": 0.0, "duration_s": 9.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-free-u00-c00.mp3" }, "en-conversation-01-base17-segment-free-u09-c00": { "id": "en-conversation-01-base17-segment-free-u09-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 12.88, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 257280, "script_start_s": 0.16, "script_end_s": 5.36, "all_asr_last_word_s": 12.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0, "source_identity_reward": 0.06964429009379543, "source_start_sample": 3840, "source_end_sample": 264480, "first_word_output_sample": 3840, "last_word_output_sample": 253440, "samples": 260640, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 7.37, "added_silence_s": 0.0, "duration_s": 5.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-free-u09-c00.mp3" }, "de-conversation-03-base17-block-timed-u01-c00": { "id": "de-conversation-03-base17-block-timed-u01-c00", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 85617, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.8, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 45, "coverage": 0.9574468085106383, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0425531914893617, "source_identity_reward": 0.3514004359353711 }, "de-conversation-03-base17-block-timed-u01-c01": { "id": "de-conversation-03-base17-block-timed-u01-c01", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 85618, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.88, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 45, "coverage": 0.9574468085106383, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06382978723404255, "source_identity_reward": 0.2331014094337398 }, "de-conversation-03-base17-block-timed-u01-c02": { "id": "de-conversation-03-base17-block-timed-u01-c02", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 85619, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 44, "coverage": 0.9361702127659575, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0851063829787234, "source_identity_reward": 0.2565613825807065 }, "de-conversation-03-base17-block-timed-u01-c03": { "id": "de-conversation-03-base17-block-timed-u01-c03", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 85620, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.8, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 44, "coverage": 0.9361702127659575, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0851063829787234, "source_identity_reward": 0.24465823396303113 }, "de-conversation-03-base17-block-timed-u01-c04": { "id": "de-conversation-03-base17-block-timed-u01-c04", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 85621, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.8, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 46, "coverage": 0.9787234042553191, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02127659574468085, "source_identity_reward": 0.22814019107510636 }, "de-conversation-03-base17-block-timed-u01-c05": { "id": "de-conversation-03-base17-block-timed-u01-c05", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 85622, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.72, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 46, "coverage": 0.9787234042553191, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0425531914893617, "source_identity_reward": 0.24929173198014704 }, "de-conversation-03-base17-block-timed-u01-c06": { "id": "de-conversation-03-base17-block-timed-u01-c06", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 85623, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.88, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 46, "coverage": 0.9787234042553191, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02127659574468085, "source_identity_reward": 0.33519438673011803 }, "de-conversation-03-base17-block-timed-u01-c07": { "id": "de-conversation-03-base17-block-timed-u01-c07", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 85624, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 16.8, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 45, "coverage": 0.9574468085106383, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0425531914893617, "source_identity_reward": 0.24809490912731266 }, "de-conversation-01-base17-segment-timed-u00-c00": { "id": "de-conversation-01-base17-segment-timed-u00-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83101, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 7, "coverage": 0.6363636363636364, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.36363636363636365, "source_identity_reward": 0.5053441297352573 }, "en-conversation-03-base17-segment-timed-u06-c00": { "id": "en-conversation-03-base17-segment-timed-u06-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75197, "plain": "References guide identity; acting directions guide expression. Neither is a magic guarantee.", "source_duration_s": 4.16, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 195840, "script_start_s": 0.0, "script_end_s": 4.08, "all_asr_last_word_s": 4.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4048361522045836, "source_start_sample": 0, "source_end_sample": 199680, "first_word_output_sample": 0, "last_word_output_sample": 195840, "samples": 203040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.23, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-timed-u06-c00.mp3" }, "de-conversation-01-base17-segment-free-u04-c00": { "id": "de-conversation-01-base17-segment-free-u04-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83165, "plain": "Genau. Wir testen vierzig Emotionsklassen und behaupten nicht, dass jede perfekt gelingt.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 11, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 345600, "script_start_s": 0.16, "script_end_s": 7.2, "all_asr_last_word_s": 7.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.7474053974747102, "source_start_sample": 3840, "source_end_sample": 352800, "first_word_output_sample": 3840, "last_word_output_sample": 341760, "samples": 348960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 22.65, "added_silence_s": 0.0, "duration_s": 7.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-free-u04-c00.mp3" }, "de-conversation-01-base17-segment-free-u00-c00": { "id": "de-conversation-01-base17-segment-free-u00-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83101, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large.", "source_duration_s": 29.28, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 8, "coverage": 0.7272727272727273, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 387840, "script_start_s": 0.24, "script_end_s": 8.08, "all_asr_last_word_s": 8.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.36363636363636365, "source_identity_reward": 0.4702567663511008, "source_start_sample": 7680, "source_end_sample": 395040, "first_word_output_sample": 3840, "last_word_output_sample": 380160, "samples": 387360, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 21.05, "added_silence_s": 0.0, "duration_s": 8.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-free-u00-c00.mp3" }, "en-conversation-04-base17-segment-timed-u03-c00": { "id": "en-conversation-04-base17-segment-timed-u03-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76149, "plain": "Thank you. I was about to ask it to fit my entire life into three seconds.", "source_duration_s": 5.36, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 249600, "script_start_s": 0.24, "script_end_s": 5.2, "all_asr_last_word_s": 5.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.39068610289112926, "source_start_sample": 7680, "source_end_sample": 256800, "first_word_output_sample": 3840, "last_word_output_sample": 241920, "samples": 249120, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-timed-u03-c00.mp3" }, "de-conversation-04-base17-segment-free-u04-c00": { "id": "de-conversation-04-base17-segment-free-u04-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86165, "plain": "Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 445440, "script_start_s": 0.0, "script_end_s": 9.28, "all_asr_last_word_s": 29.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0, "source_identity_reward": 0.32899687568063085, "source_start_sample": 0, "source_end_sample": 452640, "first_word_output_sample": 0, "last_word_output_sample": 445440, "samples": 452640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 20.57, "added_silence_s": 0.0, "duration_s": 9.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-free-u04-c00.mp3" }, "de-conversation-05-base17-segment-free-u05-c00": { "id": "de-conversation-05-base17-segment-free-u05-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87181, "plain": "Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 372480, "script_start_s": 0.0, "script_end_s": 7.76, "all_asr_last_word_s": 7.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.627920155144562, "source_start_sample": 0, "source_end_sample": 379680, "first_word_output_sample": 0, "last_word_output_sample": 372480, "samples": 379680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.41, "added_silence_s": 0.0, "duration_s": 7.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u05-c00.mp3" }, "de-conversation-05-base17-block-timed-u01-c00": { "id": "de-conversation-05-base17-block-timed-u01-c00", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 87617, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 52, "coverage": 0.9629629629629629, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 921600, "script_start_s": 0.16, "script_end_s": 19.2, "all_asr_last_word_s": 19.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.037037037037037035, "source_identity_reward": 0.23606730260854597, "source_start_sample": 3840, "source_end_sample": 925440, "first_word_output_sample": 3840, "last_word_output_sample": 917760, "samples": 924960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 19.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u01-c00.mp3" }, "de-conversation-05-base17-block-timed-u01-c01": { "id": "de-conversation-05-base17-block-timed-u01-c01", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 87618, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 917760, "script_start_s": 0.16, "script_end_s": 19.12, "all_asr_last_word_s": 19.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.30305969886512685, "source_start_sample": 3840, "source_end_sample": 924960, "first_word_output_sample": 3840, "last_word_output_sample": 913920, "samples": 921120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u01-c01.mp3" }, "de-conversation-05-base17-block-timed-u01-c02": { "id": "de-conversation-05-base17-block-timed-u01-c02", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 87619, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 51, "coverage": 0.9444444444444444, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 917760, "script_start_s": 0.16, "script_end_s": 19.12, "all_asr_last_word_s": 19.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05555555555555555, "source_identity_reward": 0.2501593141378898, "source_start_sample": 3840, "source_end_sample": 924960, "first_word_output_sample": 3840, "last_word_output_sample": 913920, "samples": 921120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u01-c02.mp3" }, "de-conversation-05-base17-block-timed-u01-c03": { "id": "de-conversation-05-base17-block-timed-u01-c03", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 87620, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.36, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 52, "coverage": 0.9629629629629629, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 921600, "script_start_s": 0.16, "script_end_s": 19.2, "all_asr_last_word_s": 19.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.037037037037037035, "source_identity_reward": 0.1949923502385267, "source_start_sample": 3840, "source_end_sample": 928800, "first_word_output_sample": 3840, "last_word_output_sample": 917760, "samples": 924960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u01-c03.mp3" }, "de-conversation-05-base17-block-timed-u01-c04": { "id": "de-conversation-05-base17-block-timed-u01-c04", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 87621, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 52, "coverage": 0.9629629629629629, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 917760, "script_start_s": 0.16, "script_end_s": 19.12, "all_asr_last_word_s": 19.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.037037037037037035, "source_identity_reward": 0.3145313040906643, "source_start_sample": 3840, "source_end_sample": 924960, "first_word_output_sample": 3840, "last_word_output_sample": 913920, "samples": 921120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u01-c04.mp3" }, "de-conversation-05-base17-block-timed-u01-c05": { "id": "de-conversation-05-base17-block-timed-u01-c05", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 87622, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 917760, "script_start_s": 0.16, "script_end_s": 19.12, "all_asr_last_word_s": 19.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.24474349926289335, "source_start_sample": 3840, "source_end_sample": 924960, "first_word_output_sample": 3840, "last_word_output_sample": 913920, "samples": 921120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u01-c05.mp3" }, "de-conversation-05-base17-block-timed-u01-c06": { "id": "de-conversation-05-base17-block-timed-u01-c06", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 87623, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 52, "coverage": 0.9629629629629629, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 917760, "script_start_s": 0.16, "script_end_s": 19.12, "all_asr_last_word_s": 19.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.037037037037037035, "source_identity_reward": 0.23532636021601286, "source_start_sample": 3840, "source_end_sample": 924960, "first_word_output_sample": 3840, "last_word_output_sample": 913920, "samples": 921120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u01-c06.mp3" }, "de-conversation-05-base17-block-timed-u01-c07": { "id": "de-conversation-05-base17-block-timed-u01-c07", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 87624, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 52, "coverage": 0.9629629629629629, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 917760, "script_start_s": 0.16, "script_end_s": 19.12, "all_asr_last_word_s": 19.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.037037037037037035, "source_identity_reward": 0.21949214132560307, "source_start_sample": 3840, "source_end_sample": 924960, "first_word_output_sample": 3840, "last_word_output_sample": 913920, "samples": 921120, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.19, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-timed-u01-c07.mp3" }, "en-conversation-02-base17-block-timed-u02-c00": { "id": "en-conversation-02-base17-block-timed-u02-c00", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 74633, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 460800, "script_start_s": 0.24, "script_end_s": 9.6, "all_asr_last_word_s": 9.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.0860960041579997, "source_start_sample": 7680, "source_end_sample": 464640, "first_word_output_sample": 3840, "last_word_output_sample": 453120, "samples": 460320, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u02-c00.mp3" }, "en-conversation-02-base17-block-timed-u02-c01": { "id": "en-conversation-02-base17-block-timed-u02-c01", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 74634, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 460800, "script_start_s": 0.16, "script_end_s": 9.6, "all_asr_last_word_s": 9.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1497638393606431, "source_start_sample": 3840, "source_end_sample": 464640, "first_word_output_sample": 3840, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u02-c01.mp3" }, "en-conversation-02-base17-block-timed-u02-c02": { "id": "en-conversation-02-base17-block-timed-u02-c02", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 74635, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 460800, "script_start_s": 0.16, "script_end_s": 9.6, "all_asr_last_word_s": 9.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.11547272627499447, "source_start_sample": 3840, "source_end_sample": 464640, "first_word_output_sample": 3840, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u02-c02.mp3" }, "en-conversation-02-base17-block-timed-u02-c03": { "id": "en-conversation-02-base17-block-timed-u02-c03", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 74636, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 460800, "script_start_s": 0.16, "script_end_s": 9.6, "all_asr_last_word_s": 9.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.14384954500181985, "source_start_sample": 3840, "source_end_sample": 464640, "first_word_output_sample": 3840, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u02-c03.mp3" }, "en-conversation-02-base17-block-timed-u02-c04": { "id": "en-conversation-02-base17-block-timed-u02-c04", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 74637, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.76, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 464640, "script_start_s": 0.24, "script_end_s": 9.68, "all_asr_last_word_s": 9.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.16096419018939864, "source_start_sample": 7680, "source_end_sample": 468480, "first_word_output_sample": 3840, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u02-c04.mp3" }, "en-conversation-02-base17-block-timed-u02-c05": { "id": "en-conversation-02-base17-block-timed-u02-c05", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 74638, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 456960, "script_start_s": 0.16, "script_end_s": 9.52, "all_asr_last_word_s": 9.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.11682213156752795, "source_start_sample": 3840, "source_end_sample": 464160, "first_word_output_sample": 3840, "last_word_output_sample": 453120, "samples": 460320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u02-c05.mp3" }, "en-conversation-02-base17-block-timed-u02-c06": { "id": "en-conversation-02-base17-block-timed-u02-c06", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 74639, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.76, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 460800, "script_start_s": 0.16, "script_end_s": 9.6, "all_asr_last_word_s": 9.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.11345891976411283, "source_start_sample": 3840, "source_end_sample": 468000, "first_word_output_sample": 3840, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 9.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u02-c06.mp3" }, "en-conversation-02-base17-block-timed-u02-c07": { "id": "en-conversation-02-base17-block-timed-u02-c07", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 74640, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 9.68, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 460800, "script_start_s": 0.16, "script_end_s": 9.6, "all_asr_last_word_s": 9.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.12960610140625212, "source_start_sample": 3840, "source_end_sample": 464640, "first_word_output_sample": 3840, "last_word_output_sample": 456960, "samples": 464160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u02-c07.mp3" }, "en-conversation-05-base17-segment-free-u03-c00": { "id": "en-conversation-05-base17-segment-free-u03-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77149, "plain": "I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 349440, "script_start_s": 0.0, "script_end_s": 7.28, "all_asr_last_word_s": 7.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2595364520700169, "source_start_sample": 0, "source_end_sample": 356640, "first_word_output_sample": 0, "last_word_output_sample": 349440, "samples": 356640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 22.57, "added_silence_s": 0.0, "duration_s": 7.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-free-u03-c00.mp3" }, "en-conversation-02-base17-segment-free-u11-c00": { "id": "en-conversation-02-base17-segment-free-u11-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 8.56, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 407040, "script_start_s": 0.24, "script_end_s": 8.48, "all_asr_last_word_s": 8.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1378466114225576, "source_start_sample": 7680, "source_end_sample": 410880, "first_word_output_sample": 3840, "last_word_output_sample": 399360, "samples": 406560, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 8.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-free-u11-c00.mp3" }, "de-conversation-01-base17-segment-timed-u09-c00": { "id": "de-conversation-01-base17-segment-timed-u09-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 230400, "script_start_s": 0.0, "script_end_s": 4.8, "all_asr_last_word_s": 4.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.27382149962186175, "source_start_sample": 0, "source_end_sample": 237600, "first_word_output_sample": 0, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-timed-u09-c00.mp3" }, "en-conversation-01-base17-segment-free-u06-c00": { "id": "en-conversation-01-base17-segment-free-u06-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73197, "plain": "The new training mixes real conversations with acting examples to retain both formats.", "source_duration_s": 8.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 276480, "script_start_s": 0.0, "script_end_s": 5.76, "all_asr_last_word_s": 5.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.36020231029384797, "source_start_sample": 0, "source_end_sample": 283680, "first_word_output_sample": 0, "last_word_output_sample": 276480, "samples": 283680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.81, "added_silence_s": 0.0, "duration_s": 5.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-free-u06-c00.mp3" }, "en-conversation-02-base17-block-timed-u00-c00": { "id": "en-conversation-02-base17-block-timed-u00-c00", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 74601, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.08, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1052160, "script_start_s": 0.16, "script_end_s": 21.92, "all_asr_last_word_s": 21.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029850746268656716, "source_identity_reward": 0.143075345637651, "source_start_sample": 3840, "source_end_sample": 1059360, "first_word_output_sample": 3840, "last_word_output_sample": 1048320, "samples": 1055520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u00-c00.mp3" }, "en-conversation-02-base17-block-timed-u00-c01": { "id": "en-conversation-02-base17-block-timed-u00-c01", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 74602, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.08, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1052160, "script_start_s": 0.16, "script_end_s": 21.92, "all_asr_last_word_s": 21.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.14402738399671078, "source_start_sample": 3840, "source_end_sample": 1059360, "first_word_output_sample": 3840, "last_word_output_sample": 1048320, "samples": 1055520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u00-c01.mp3" }, "en-conversation-02-base17-block-timed-u00-c02": { "id": "en-conversation-02-base17-block-timed-u00-c02", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 74603, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.08, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1056000, "script_start_s": 0.16, "script_end_s": 22.0, "all_asr_last_word_s": 22.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029850746268656716, "source_identity_reward": 0.12823513699533326, "source_start_sample": 3840, "source_end_sample": 1059840, "first_word_output_sample": 3840, "last_word_output_sample": 1052160, "samples": 1059360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 22.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u00-c02.mp3" }, "en-conversation-02-base17-block-timed-u00-c03": { "id": "en-conversation-02-base17-block-timed-u00-c03", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 74604, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.08, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1052160, "script_start_s": 0.16, "script_end_s": 21.92, "all_asr_last_word_s": 21.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029850746268656716, "source_identity_reward": 0.16096756260495645, "source_start_sample": 3840, "source_end_sample": 1059360, "first_word_output_sample": 3840, "last_word_output_sample": 1048320, "samples": 1055520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u00-c03.mp3" }, "en-conversation-02-base17-block-timed-u00-c04": { "id": "en-conversation-02-base17-block-timed-u00-c04", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 74605, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.08, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1052160, "script_start_s": 0.24, "script_end_s": 21.92, "all_asr_last_word_s": 21.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.11680892254618915, "source_start_sample": 7680, "source_end_sample": 1059360, "first_word_output_sample": 3840, "last_word_output_sample": 1044480, "samples": 1051680, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u00-c04.mp3" }, "en-conversation-02-base17-block-timed-u00-c05": { "id": "en-conversation-02-base17-block-timed-u00-c05", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 74606, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.08, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1052160, "script_start_s": 0.16, "script_end_s": 21.92, "all_asr_last_word_s": 21.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.15202669334311827, "source_start_sample": 3840, "source_end_sample": 1059360, "first_word_output_sample": 3840, "last_word_output_sample": 1048320, "samples": 1055520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u00-c05.mp3" }, "en-conversation-02-base17-block-timed-u00-c06": { "id": "en-conversation-02-base17-block-timed-u00-c06", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 74607, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.4, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029850746268656716, "source_identity_reward": 0.10770637364140548 }, "en-conversation-02-base17-block-timed-u00-c07": { "id": "en-conversation-02-base17-block-timed-u00-c07", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 74608, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model. Please tell me the tiny studio has a very dramatic red recording light. Of course. Humaneness Voice Large supplies the foundation; podcast fine-tuning teaches dialogue. And does the second host interrupt with healthy skepticism, or just politely agree? Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 22.08, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1056000, "script_start_s": 0.16, "script_end_s": 22.0, "all_asr_last_word_s": 22.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029850746268656716, "source_identity_reward": 0.10179965603356102, "source_start_sample": 3840, "source_end_sample": 1059840, "first_word_output_sample": 3840, "last_word_output_sample": 1052160, "samples": 1059360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 22.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-timed-u00-c07.mp3" }, "en-conversation-04-base17-segment-timed-u02-c00": { "id": "en-conversation-04-base17-segment-timed-u02-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76133, "plain": "No. The codec advances in eighty-millisecond frames, and requested timing can still miss.", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 13, "coverage": 0.9285714285714286, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 207360, "script_start_s": 0.0, "script_end_s": 4.32, "all_asr_last_word_s": 4.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07142857142857142, "source_identity_reward": 0.44124593779592935, "source_start_sample": 0, "source_end_sample": 211200, "first_word_output_sample": 0, "last_word_output_sample": 207360, "samples": 214560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-timed-u02-c00.mp3" }, "de-conversation-03-base17-segment-free-u00-c00": { "id": "de-conversation-03-base17-segment-free-u00-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85101, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können.", "source_duration_s": 11.84, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 560640, "script_start_s": 0.16, "script_end_s": 11.68, "all_asr_last_word_s": 11.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7133577563355251, "source_start_sample": 3840, "source_end_sample": 567840, "first_word_output_sample": 3840, "last_word_output_sample": 556800, "samples": 564000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 11.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-free-u00-c00.mp3" }, "de-conversation-02-base17-segment-free-u00-c00": { "id": "de-conversation-02-base17-segment-free-u00-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84101, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 11, "coverage": 0.8461538461538461, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.15384615384615385, "source_identity_reward": 0.5965932230023745 }, "en-conversation-01-base17-segment-timed-u11-c00": { "id": "en-conversation-01-base17-segment-timed-u11-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 4.96, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 234240, "script_start_s": 0.24, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.072827713934798, "source_start_sample": 7680, "source_end_sample": 238080, "first_word_output_sample": 3840, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-timed-u11-c00.mp3" }, "en-conversation-05-base17-block-free-u00-c00": { "id": "en-conversation-05-base17-block-free-u00-c00", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 77601, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1098240, "script_start_s": 0.16, "script_end_s": 22.88, "all_asr_last_word_s": 59.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03636363636363636, "source_identity_reward": 0.26615106994346466, "source_start_sample": 3840, "source_end_sample": 1105440, "first_word_output_sample": 3840, "last_word_output_sample": 1094400, "samples": 1101600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 36.97, "added_silence_s": 0.0, "duration_s": 22.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u00-c00.mp3" }, "en-conversation-05-base17-block-free-u00-c01": { "id": "en-conversation-05-base17-block-free-u00-c01", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 77602, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 22.24, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1056000, "script_start_s": 0.16, "script_end_s": 22.0, "all_asr_last_word_s": 22.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.21834816506469923, "source_start_sample": 3840, "source_end_sample": 1063200, "first_word_output_sample": 3840, "last_word_output_sample": 1052160, "samples": 1059360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 22.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u00-c01.mp3" }, "en-conversation-05-base17-block-free-u00-c02": { "id": "en-conversation-05-base17-block-free-u00-c02", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 77603, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 22.88, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 53, "coverage": 0.9636363636363636, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1090560, "script_start_s": 0.16, "script_end_s": 22.72, "all_asr_last_word_s": 22.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03636363636363636, "source_identity_reward": 0.23976866145159226, "source_start_sample": 3840, "source_end_sample": 1097760, "first_word_output_sample": 3840, "last_word_output_sample": 1086720, "samples": 1093920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u00-c02.mp3" }, "en-conversation-05-base17-block-free-u00-c03": { "id": "en-conversation-05-base17-block-free-u00-c03", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 77604, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 961920, "script_start_s": 0.16, "script_end_s": 20.04, "all_asr_last_word_s": 37.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.14545454545454545, "source_identity_reward": 0.1702433229058168, "source_start_sample": 3840, "source_end_sample": 969120, "first_word_output_sample": 3840, "last_word_output_sample": 958080, "samples": 965280, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 39.81, "added_silence_s": 0.0, "duration_s": 20.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u00-c03.mp3" }, "en-conversation-05-base17-block-free-u00-c04": { "id": "en-conversation-05-base17-block-free-u00-c04", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 77605, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 20.72, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 979200, "script_start_s": 0.16, "script_end_s": 20.4, "all_asr_last_word_s": 20.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1480647235221159, "source_start_sample": 3840, "source_end_sample": 986400, "first_word_output_sample": 3840, "last_word_output_sample": 975360, "samples": 982560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.17, "added_silence_s": 0.0, "duration_s": 20.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u00-c04.mp3" }, "en-conversation-05-base17-block-free-u00-c05": { "id": "en-conversation-05-base17-block-free-u00-c05", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 77606, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 23.04, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 53, "coverage": 0.9636363636363636, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1075200, "script_start_s": 0.08, "script_end_s": 22.4, "all_asr_last_word_s": 22.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03636363636363636, "source_identity_reward": 0.1263363425734645, "source_start_sample": 0, "source_end_sample": 1082400, "first_word_output_sample": 3840, "last_word_output_sample": 1075200, "samples": 1082400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.49, "added_silence_s": 0.0, "duration_s": 22.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u00-c05.mp3" }, "en-conversation-05-base17-block-free-u00-c06": { "id": "en-conversation-05-base17-block-free-u00-c06", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 77607, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 23.52, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 52, "coverage": 0.9454545454545454, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05454545454545454, "source_identity_reward": 0.21699410409092837 }, "en-conversation-05-base17-block-free-u00-c07": { "id": "en-conversation-05-base17-block-free-u00-c07", "case": "en-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 77608, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 912000, "script_start_s": 0.16, "script_end_s": 19.0, "all_asr_last_word_s": 57.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03636363636363636, "source_identity_reward": 0.25355416116425433, "source_start_sample": 3840, "source_end_sample": 919200, "first_word_output_sample": 3840, "last_word_output_sample": 908160, "samples": 915360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.85, "added_silence_s": 0.0, "duration_s": 19.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-block-free-u00-c07.mp3" }, "de-conversation-02-base17-block-timed-u01-c00": { "id": "de-conversation-02-base17-block-timed-u01-c00", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 84617, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.92, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1136640, "script_start_s": 0.16, "script_end_s": 23.68, "all_asr_last_word_s": 23.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.22247623849849046, "source_start_sample": 3840, "source_end_sample": 1143840, "first_word_output_sample": 3840, "last_word_output_sample": 1132800, "samples": 1140000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 23.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u01-c00.mp3" }, "de-conversation-02-base17-block-timed-u01-c01": { "id": "de-conversation-02-base17-block-timed-u01-c01", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 84618, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.84, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1136640, "script_start_s": 0.24, "script_end_s": 23.68, "all_asr_last_word_s": 23.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.39565250803328283, "source_start_sample": 7680, "source_end_sample": 1143840, "first_word_output_sample": 3840, "last_word_output_sample": 1128960, "samples": 1136160, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u01-c01.mp3" }, "de-conversation-02-base17-block-timed-u01-c02": { "id": "de-conversation-02-base17-block-timed-u01-c02", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 84619, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.92, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1140480, "script_start_s": 0.16, "script_end_s": 23.76, "all_asr_last_word_s": 23.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029850746268656716, "source_identity_reward": 0.19810047170452308, "source_start_sample": 3840, "source_end_sample": 1147680, "first_word_output_sample": 3840, "last_word_output_sample": 1136640, "samples": 1143840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u01-c02.mp3" }, "de-conversation-02-base17-block-timed-u01-c03": { "id": "de-conversation-02-base17-block-timed-u01-c03", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 84620, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.84, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1132800, "script_start_s": 0.16, "script_end_s": 23.6, "all_asr_last_word_s": 23.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.2228608671840479, "source_start_sample": 3840, "source_end_sample": 1140000, "first_word_output_sample": 3840, "last_word_output_sample": 1128960, "samples": 1136160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 23.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u01-c03.mp3" }, "de-conversation-02-base17-block-timed-u01-c04": { "id": "de-conversation-02-base17-block-timed-u01-c04", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 84621, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.92, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1140480, "script_start_s": 0.16, "script_end_s": 23.76, "all_asr_last_word_s": 23.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.1985602445181728, "source_start_sample": 3840, "source_end_sample": 1147680, "first_word_output_sample": 3840, "last_word_output_sample": 1136640, "samples": 1143840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u01-c04.mp3" }, "de-conversation-02-base17-block-timed-u01-c05": { "id": "de-conversation-02-base17-block-timed-u01-c05", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 84622, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.92, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1144320, "script_start_s": 0.16, "script_end_s": 23.84, "all_asr_last_word_s": 23.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.23971217132782258, "source_start_sample": 3840, "source_end_sample": 1148160, "first_word_output_sample": 3840, "last_word_output_sample": 1140480, "samples": 1147680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u01-c05.mp3" }, "de-conversation-02-base17-block-timed-u01-c06": { "id": "de-conversation-02-base17-block-timed-u01-c06", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 84623, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.84, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 66, "coverage": 0.9850746268656716, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1132800, "script_start_s": 0.16, "script_end_s": 23.6, "all_asr_last_word_s": 23.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014925373134328358, "source_identity_reward": 0.24305485212497083, "source_start_sample": 3840, "source_end_sample": 1140000, "first_word_output_sample": 3840, "last_word_output_sample": 1128960, "samples": 1136160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 23.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u01-c06.mp3" }, "de-conversation-02-base17-block-timed-u01-c07": { "id": "de-conversation-02-base17-block-timed-u01-c07", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 84624, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören. Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen. Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.92, "source_frame_cap": false, "alignment": { "expected_tokens": 67, "matched_tokens": 65, "coverage": 0.9701492537313433, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1140480, "script_start_s": 0.16, "script_end_s": 23.76, "all_asr_last_word_s": 23.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.029850746268656716, "source_identity_reward": 0.23668881862482868, "source_start_sample": 3840, "source_end_sample": 1147680, "first_word_output_sample": 3840, "last_word_output_sample": 1136640, "samples": 1143840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u01-c07.mp3" }, "de-conversation-04-base17-segment-timed-u00-c00": { "id": "de-conversation-04-base17-segment-timed-u00-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86101, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben.", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 203520, "script_start_s": 0.0, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7318195879115951, "source_start_sample": 0, "source_end_sample": 207360, "first_word_output_sample": 0, "last_word_output_sample": 203520, "samples": 210720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-timed-u00-c00.mp3" }, "en-conversation-03-base17-segment-timed-u03-c00": { "id": "en-conversation-03-base17-segment-timed-u03-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75149, "plain": "I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 218880, "script_start_s": 0.24, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2736057116939259, "source_start_sample": 7680, "source_end_sample": 222720, "first_word_output_sample": 3840, "last_word_output_sample": 211200, "samples": 218400, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-timed-u03-c00.mp3" }, "en-conversation-05-base17-segment-free-u13-c00": { "id": "en-conversation-05-base17-segment-free-u13-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77309, "plain": "And the second host is allowed to say, hold on, explain that again, I have not followed you.", "source_duration_s": 8.96, "source_frame_cap": false, "alignment": { "expected_tokens": 18, "matched_tokens": 18, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 418560, "script_start_s": 0.16, "script_end_s": 8.72, "all_asr_last_word_s": 8.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u13-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.11247239834782893, "source_start_sample": 3840, "source_end_sample": 425760, "first_word_output_sample": 3840, "last_word_output_sample": 414720, "samples": 421920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 8.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-free-u13-c00.mp3" }, "de-conversation-05-base17-segment-timed-u05-c00": { "id": "de-conversation-05-base17-segment-timed-u05-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87181, "plain": "Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 11, "coverage": 0.8461538461538461, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.15384615384615385, "source_identity_reward": 0.277640241023246 }, "de-conversation-02-base17-segment-free-u06-c00": { "id": "de-conversation-02-base17-segment-free-u06-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84197, "plain": "Deshalb stehen die Referenzclips direkt neben den Ausgaben. Du kannst selbst vergleichen.", "source_duration_s": 12.4, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 11, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 268800, "script_start_s": 0.0, "script_end_s": 5.6, "all_asr_last_word_s": 5.6000000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.6315086439829838, "source_start_sample": 0, "source_end_sample": 276000, "first_word_output_sample": 0, "last_word_output_sample": 268800, "samples": 276000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 6.65, "added_silence_s": 0.0, "duration_s": 5.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-free-u06-c00.mp3" }, "de-conversation-01-base17-segment-free-u07-c00": { "id": "de-conversation-01-base17-segment-free-u07-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83213, "plain": "Na gut. Zeig mir diese kleine Pause, bevor jemand widerwillig zugibt, beeindruckt zu sein.", "source_duration_s": 7.84, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 13, "coverage": 0.9285714285714286, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 337920, "script_start_s": 0.24, "script_end_s": 7.04, "all_asr_last_word_s": 7.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07142857142857142, "source_identity_reward": 0.47615365262787024, "source_start_sample": 7680, "source_end_sample": 345120, "first_word_output_sample": 3840, "last_word_output_sample": 330240, "samples": 337440, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.65, "added_silence_s": 0.0, "duration_s": 7.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-free-u07-c00.mp3" }, "en-conversation-03-base17-segment-free-u04-c00": { "id": "en-conversation-03-base17-segment-free-u04-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75165, "plain": "We test that directly, turn by turn, rather than trusting the first sentence.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 261120, "script_start_s": 0.0, "script_end_s": 5.44, "all_asr_last_word_s": 5.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.303491401677902, "source_start_sample": 0, "source_end_sample": 268320, "first_word_output_sample": 0, "last_word_output_sample": 261120, "samples": 268320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.41, "added_silence_s": 0.0, "duration_s": 5.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-free-u04-c00.mp3" }, "en-conversation-03-base17-block-timed-u01-c00": { "id": "en-conversation-03-base17-block-timed-u01-c00", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 75617, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 810240, "script_start_s": 0.24, "script_end_s": 16.88, "all_asr_last_word_s": 16.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.12616936949837634, "source_start_sample": 7680, "source_end_sample": 814080, "first_word_output_sample": 3840, "last_word_output_sample": 802560, "samples": 809760, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u01-c00.mp3" }, "en-conversation-03-base17-block-timed-u01-c01": { "id": "en-conversation-03-base17-block-timed-u01-c01", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 75618, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 810240, "script_start_s": 0.16, "script_end_s": 16.88, "all_asr_last_word_s": 16.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1808971103211255, "source_start_sample": 3840, "source_end_sample": 814080, "first_word_output_sample": 3840, "last_word_output_sample": 806400, "samples": 813600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u01-c01.mp3" }, "en-conversation-03-base17-block-timed-u01-c02": { "id": "en-conversation-03-base17-block-timed-u01-c02", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 75619, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 50, "coverage": 0.9803921568627451, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 806400, "script_start_s": 0.0, "script_end_s": 16.8, "all_asr_last_word_s": 16.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0196078431372549, "source_identity_reward": 0.15184797927687782, "source_start_sample": 0, "source_end_sample": 813600, "first_word_output_sample": 0, "last_word_output_sample": 806400, "samples": 813600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u01-c02.mp3" }, "en-conversation-03-base17-block-timed-u01-c03": { "id": "en-conversation-03-base17-block-timed-u01-c03", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 75620, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 806400, "script_start_s": 0.16, "script_end_s": 16.8, "all_asr_last_word_s": 16.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.153346827126237, "source_start_sample": 3840, "source_end_sample": 813600, "first_word_output_sample": 3840, "last_word_output_sample": 802560, "samples": 809760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u01-c03.mp3" }, "en-conversation-03-base17-block-timed-u01-c04": { "id": "en-conversation-03-base17-block-timed-u01-c04", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 75621, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 810240, "script_start_s": 0.16, "script_end_s": 16.88, "all_asr_last_word_s": 16.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.21225442072375286, "source_start_sample": 3840, "source_end_sample": 814080, "first_word_output_sample": 3840, "last_word_output_sample": 806400, "samples": 813600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u01-c04.mp3" }, "en-conversation-03-base17-block-timed-u01-c05": { "id": "en-conversation-03-base17-block-timed-u01-c05", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 75622, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 50, "coverage": 0.9803921568627451, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 810240, "script_start_s": 0.16, "script_end_s": 16.88, "all_asr_last_word_s": 16.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0196078431372549, "source_identity_reward": 0.1925986592831104, "source_start_sample": 3840, "source_end_sample": 814080, "first_word_output_sample": 3840, "last_word_output_sample": 806400, "samples": 813600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u01-c05.mp3" }, "en-conversation-03-base17-block-timed-u01-c06": { "id": "en-conversation-03-base17-block-timed-u01-c06", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 75623, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 51, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 806400, "script_start_s": 0.0, "script_end_s": 16.8, "all_asr_last_word_s": 16.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.132845763674948, "source_start_sample": 0, "source_end_sample": 813600, "first_word_output_sample": 0, "last_word_output_sample": 806400, "samples": 813600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 16.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u01-c06.mp3" }, "en-conversation-03-base17-block-timed-u01-c07": { "id": "en-conversation-03-base17-block-timed-u01-c07", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 75624, "plain": "We test that directly, turn by turn, rather than trusting the first sentence. Good. I want the delighted host to remain the same person when she becomes doubtful. References guide identity; acting directions guide expression. Neither is a magic guarantee. Somehow the honest explanation is more exciting than the exaggerated one.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 51, "matched_tokens": 50, "coverage": 0.9803921568627451, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 810240, "script_start_s": 0.16, "script_end_s": 16.88, "all_asr_last_word_s": 16.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0392156862745098, "source_identity_reward": 0.1841248158973747, "source_start_sample": 3840, "source_end_sample": 814080, "first_word_output_sample": 3840, "last_word_output_sample": 806400, "samples": 813600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 16.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u01-c07.mp3" }, "en-conversation-05-base17-segment-free-u01-c00": { "id": "en-conversation-05-base17-segment-free-u01-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77117, "plain": "Same model family, different voices, and a different emotional rhythm in each language?", "source_duration_s": 11.2, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 276480, "script_start_s": 0.16, "script_end_s": 5.76, "all_asr_last_word_s": 5.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.31904376661894446, "source_start_sample": 3840, "source_end_sample": 283680, "first_word_output_sample": 3840, "last_word_output_sample": 272640, "samples": 279840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 5.29, "added_silence_s": 0.0, "duration_s": 5.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-free-u01-c00.mp3" }, "en-conversation-01-base17-segment-free-u00-c00": { "id": "en-conversation-01-base17-segment-free-u00-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73101, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 12, "coverage": 0.8571428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 307200, "script_start_s": 0.16, "script_end_s": 6.4, "all_asr_last_word_s": 29.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.7857142857142858, "source_identity_reward": 0.20569237499480747, "source_start_sample": 3840, "source_end_sample": 314400, "first_word_output_sample": 3840, "last_word_output_sample": 303360, "samples": 310560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.45, "added_silence_s": 0.0, "duration_s": 6.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-free-u00-c00.mp3" }, "de-conversation-05-base17-segment-free-u10-c00": { "id": "de-conversation-05-base17-segment-free-u10-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 299520, "script_start_s": 0.0, "script_end_s": 6.24, "all_asr_last_word_s": 6.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6647944699016252, "source_start_sample": 0, "source_end_sample": 306720, "first_word_output_sample": 0, "last_word_output_sample": 299520, "samples": 306720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.61, "added_silence_s": 0.0, "duration_s": 6.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u10-c00.mp3" }, "de-conversation-02-base17-block-timed-u00-c00": { "id": "de-conversation-02-base17-block-timed-u00-c00", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 84601, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.52, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 56, "coverage": 0.9333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1029120, "script_start_s": 0.16, "script_end_s": 21.44, "all_asr_last_word_s": 21.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08333333333333333, "source_identity_reward": 0.2352604969181598, "source_start_sample": 3840, "source_end_sample": 1032960, "first_word_output_sample": 3840, "last_word_output_sample": 1025280, "samples": 1032480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u00-c00.mp3" }, "de-conversation-02-base17-block-timed-u00-c01": { "id": "de-conversation-02-base17-block-timed-u00-c01", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 84602, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.52, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 59, "coverage": 0.9833333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1029120, "script_start_s": 0.08, "script_end_s": 21.44, "all_asr_last_word_s": 21.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03333333333333333, "source_identity_reward": 0.29388688284728537, "source_start_sample": 0, "source_end_sample": 1032960, "first_word_output_sample": 3840, "last_word_output_sample": 1029120, "samples": 1036320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u00-c01.mp3" }, "de-conversation-02-base17-block-timed-u00-c02": { "id": "de-conversation-02-base17-block-timed-u00-c02", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 84603, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.68, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 54, "coverage": 0.9, "first_script_word_verified": false, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.11666666666666667, "source_identity_reward": 0.219406013365614 }, "de-conversation-02-base17-block-timed-u00-c03": { "id": "de-conversation-02-base17-block-timed-u00-c03", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 84604, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.6, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 54, "coverage": 0.9, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1032960, "script_start_s": 0.0, "script_end_s": 21.52, "all_asr_last_word_s": 21.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.11666666666666667, "source_identity_reward": 0.25130145911194346, "source_start_sample": 0, "source_end_sample": 1036800, "first_word_output_sample": 0, "last_word_output_sample": 1032960, "samples": 1040160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u00-c03.mp3" }, "de-conversation-02-base17-block-timed-u00-c04": { "id": "de-conversation-02-base17-block-timed-u00-c04", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 84605, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.6, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 57, "coverage": 0.95, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06666666666666667, "source_identity_reward": 0.2518127598832048 }, "de-conversation-02-base17-block-timed-u00-c05": { "id": "de-conversation-02-base17-block-timed-u00-c05", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 84606, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.52, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 55, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1029120, "script_start_s": 0.16, "script_end_s": 21.44, "all_asr_last_word_s": 21.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1, "source_identity_reward": 0.24530263696405025, "source_start_sample": 3840, "source_end_sample": 1032960, "first_word_output_sample": 3840, "last_word_output_sample": 1025280, "samples": 1032480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u00-c05.mp3" }, "de-conversation-02-base17-block-timed-u00-c06": { "id": "de-conversation-02-base17-block-timed-u00-c06", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 84607, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.52, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 55, "coverage": 0.9166666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1029120, "script_start_s": 0.0, "script_end_s": 21.44, "all_asr_last_word_s": 21.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1, "source_identity_reward": 0.2578098404333807, "source_start_sample": 0, "source_end_sample": 1032960, "first_word_output_sample": 0, "last_word_output_sample": 1029120, "samples": 1036320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u00-c06.mp3" }, "de-conversation-02-base17-block-timed-u00-c07": { "id": "de-conversation-02-base17-block-timed-u00-c07", "case": "de-conversation-02", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 84608, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken. Bitte sag mir, dass dort eine völlig übertriebene rote Aufnahmelampe leuchtet. Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog. Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen? Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 21.52, "source_frame_cap": false, "alignment": { "expected_tokens": 60, "matched_tokens": 57, "coverage": 0.95, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1029120, "script_start_s": 0.0, "script_end_s": 21.44, "all_asr_last_word_s": 21.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06666666666666667, "source_identity_reward": 0.28666426053478683, "source_start_sample": 0, "source_end_sample": 1032960, "first_word_output_sample": 0, "last_word_output_sample": 1029120, "samples": 1036320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-block-timed-u00-c07.mp3" }, "de-conversation-04-base17-block-free-u02-c00": { "id": "de-conversation-04-base17-block-free-u02-c00", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 86633, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 545280, "script_start_s": 0.08, "script_end_s": 11.36, "all_asr_last_word_s": 14.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1724137931034483, "source_identity_reward": 0.1946875208284064, "source_start_sample": 0, "source_end_sample": 552480, "first_word_output_sample": 3840, "last_word_output_sample": 545280, "samples": 552480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 7.77, "added_silence_s": 0.0, "duration_s": 11.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u02-c00.mp3" }, "de-conversation-04-base17-block-free-u02-c01": { "id": "de-conversation-04-base17-block-free-u02-c01", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 86634, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 585600, "script_start_s": 0.0, "script_end_s": 12.2, "all_asr_last_word_s": 12.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3334892976658691, "source_start_sample": 0, "source_end_sample": 592800, "first_word_output_sample": 0, "last_word_output_sample": 585600, "samples": 592800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 47.65, "added_silence_s": 0.0, "duration_s": 12.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u02-c01.mp3" }, "de-conversation-04-base17-block-free-u02-c02": { "id": "de-conversation-04-base17-block-free-u02-c02", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 86635, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 587520, "script_start_s": 0.0, "script_end_s": 12.24, "all_asr_last_word_s": 32.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.10344827586206896, "source_identity_reward": 0.2195721642578159, "source_start_sample": 0, "source_end_sample": 594720, "first_word_output_sample": 0, "last_word_output_sample": 587520, "samples": 594720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 47.61, "added_silence_s": 0.0, "duration_s": 12.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u02-c02.mp3" }, "de-conversation-04-base17-block-free-u02-c03": { "id": "de-conversation-04-base17-block-free-u02-c03", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 86636, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 40.64, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 514560, "script_start_s": 0.0, "script_end_s": 10.72, "all_asr_last_word_s": 33.160000000000004 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5172413793103449, "source_identity_reward": 0.163317045609155, "source_start_sample": 0, "source_end_sample": 521760, "first_word_output_sample": 0, "last_word_output_sample": 514560, "samples": 521760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 29.77, "added_silence_s": 0.0, "duration_s": 10.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u02-c03.mp3" }, "de-conversation-04-base17-block-free-u02-c04": { "id": "de-conversation-04-base17-block-free-u02-c04", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 86637, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 522240, "script_start_s": 0.0, "script_end_s": 10.88, "all_asr_last_word_s": 46.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3103448275862069, "source_identity_reward": 0.23529446866764822, "source_start_sample": 0, "source_end_sample": 529440, "first_word_output_sample": 0, "last_word_output_sample": 522240, "samples": 529440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 48.97, "added_silence_s": 0.0, "duration_s": 11.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u02-c04.mp3" }, "de-conversation-04-base17-block-free-u02-c05": { "id": "de-conversation-04-base17-block-free-u02-c05", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 86638, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 18.16, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 591360, "script_start_s": 0.0, "script_end_s": 12.32, "all_asr_last_word_s": 12.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.305474647269912, "source_start_sample": 0, "source_end_sample": 598560, "first_word_output_sample": 0, "last_word_output_sample": 591360, "samples": 598560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 5.69, "added_silence_s": 0.0, "duration_s": 12.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u02-c05.mp3" }, "de-conversation-04-base17-block-free-u02-c06": { "id": "de-conversation-04-base17-block-free-u02-c06", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 86639, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 14.0, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 510720, "script_start_s": 0.0, "script_end_s": 10.64, "all_asr_last_word_s": 10.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.289601173607214, "source_start_sample": 0, "source_end_sample": 517920, "first_word_output_sample": 0, "last_word_output_sample": 510720, "samples": 517920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 3.21, "added_silence_s": 0.0, "duration_s": 10.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u02-c06.mp3" }, "de-conversation-04-base17-block-free-u02-c07": { "id": "de-conversation-04-base17-block-free-u02-c07", "case": "de-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 86640, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 16.24, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 579840, "script_start_s": 0.16, "script_end_s": 12.08, "all_asr_last_word_s": 15.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06896551724137931, "source_identity_reward": 0.2712350802421029, "source_start_sample": 3840, "source_end_sample": 587040, "first_word_output_sample": 3840, "last_word_output_sample": 576000, "samples": 583200, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 4.01, "added_silence_s": 0.0, "duration_s": 12.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-block-free-u02-c07.mp3" }, "en-conversation-03-base17-segment-free-u06-c00": { "id": "en-conversation-03-base17-segment-free-u06-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75197, "plain": "References guide identity; acting directions guide expression. Neither is a magic guarantee.", "source_duration_s": 24.4, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 407040, "script_start_s": 0.16, "script_end_s": 8.48, "all_asr_last_word_s": 12.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.6666666666666666, "source_identity_reward": 0.2375263727984534, "source_start_sample": 3840, "source_end_sample": 414240, "first_word_output_sample": 3840, "last_word_output_sample": 403200, "samples": 410400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 15.77, "added_silence_s": 0.0, "duration_s": 8.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-free-u06-c00.mp3" }, "en-conversation-04-base17-block-free-u00-c00": { "id": "en-conversation-04-base17-block-free-u00-c00", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 76601, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 20.48, "source_frame_cap": false, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 971520, "script_start_s": 0.0, "script_end_s": 20.24, "all_asr_last_word_s": 20.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.20616399182442047, "source_start_sample": 0, "source_end_sample": 978720, "first_word_output_sample": 0, "last_word_output_sample": 971520, "samples": 978720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 20.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u00-c00.mp3" }, "en-conversation-04-base17-block-free-u00-c01": { "id": "en-conversation-04-base17-block-free-u00-c01", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 76602, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1002240, "script_start_s": 0.16, "script_end_s": 20.88, "all_asr_last_word_s": 20.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.2348902585170097, "source_start_sample": 3840, "source_end_sample": 1009440, "first_word_output_sample": 3840, "last_word_output_sample": 998400, "samples": 1005600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 38.97, "added_silence_s": 0.0, "duration_s": 20.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u00-c01.mp3" }, "en-conversation-04-base17-block-free-u00-c02": { "id": "en-conversation-04-base17-block-free-u00-c02", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 76603, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 68, "matched_tokens": 66, "coverage": 0.9705882352941176, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 55.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.27941176470588236, "source_identity_reward": 0.11739847516178394, "source_start_sample": 0, "source_end_sample": 963360, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 39.93, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u00-c02.mp3" }, "en-conversation-04-base17-block-free-u00-c03": { "id": "en-conversation-04-base17-block-free-u00-c03", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 76604, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 983040, "script_start_s": 0.0, "script_end_s": 20.48, "all_asr_last_word_s": 20.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.1190271028884468, "source_start_sample": 0, "source_end_sample": 990240, "first_word_output_sample": 0, "last_word_output_sample": 983040, "samples": 990240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 39.37, "added_silence_s": 0.0, "duration_s": 20.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u00-c03.mp3" }, "en-conversation-04-base17-block-free-u00-c04": { "id": "en-conversation-04-base17-block-free-u00-c04", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 76605, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 986880, "script_start_s": 0.16, "script_end_s": 20.56, "all_asr_last_word_s": 20.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.2548271700106253, "source_start_sample": 3840, "source_end_sample": 994080, "first_word_output_sample": 3840, "last_word_output_sample": 983040, "samples": 990240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 39.29, "added_silence_s": 0.0, "duration_s": 20.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u00-c04.mp3" }, "en-conversation-04-base17-block-free-u00-c05": { "id": "en-conversation-04-base17-block-free-u00-c05", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 76606, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 979200, "script_start_s": 0.16, "script_end_s": 20.4, "all_asr_last_word_s": 20.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014705882352941176, "source_identity_reward": 0.1783342861178546, "source_start_sample": 3840, "source_end_sample": 986400, "first_word_output_sample": 3840, "last_word_output_sample": 975360, "samples": 982560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 39.45, "added_silence_s": 0.0, "duration_s": 20.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u00-c05.mp3" }, "en-conversation-04-base17-block-free-u00-c06": { "id": "en-conversation-04-base17-block-free-u00-c06", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 76607, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 990720, "script_start_s": 0.16, "script_end_s": 20.64, "all_asr_last_word_s": 30.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2647058823529412, "source_identity_reward": 0.12302716521552205, "source_start_sample": 3840, "source_end_sample": 997920, "first_word_output_sample": 3840, "last_word_output_sample": 986880, "samples": 994080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 39.21, "added_silence_s": 0.0, "duration_s": 20.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u00-c06.mp3" }, "en-conversation-04-base17-block-free-u00-c07": { "id": "en-conversation-04-base17-block-free-u00-c07", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 76608, "plain": "I can write a requested length next to an individual sentence in the script. You are kidding. Do you mean it lands on an exact millisecond? No. The codec advances in eighty-millisecond frames, and requested timing can still miss. Thank you. I was about to ask it to fit my entire life into three seconds. We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 68, "matched_tokens": 67, "coverage": 0.9852941176470589, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1017600, "script_start_s": 0.0, "script_end_s": 21.2, "all_asr_last_word_s": 45.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.08823529411764706, "source_identity_reward": 0.1431954932837617, "source_start_sample": 0, "source_end_sample": 1024800, "first_word_output_sample": 0, "last_word_output_sample": 1017600, "samples": 1024800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.65, "added_silence_s": 0.0, "duration_s": 21.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-free-u00-c07.mp3" }, "de-conversation-03-base17-block-free-u02-c00": { "id": "de-conversation-03-base17-block-free-u02-c00", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 85633, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 24.72, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 921600, "script_start_s": 0.0, "script_end_s": 19.2, "all_asr_last_word_s": 23.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.21428571428571427, "source_identity_reward": 0.207153644345547, "source_start_sample": 0, "source_end_sample": 928800, "first_word_output_sample": 0, "last_word_output_sample": 921600, "samples": 928800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 5.37, "added_silence_s": 0.0, "duration_s": 19.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u02-c00.mp3" }, "de-conversation-03-base17-block-free-u02-c01": { "id": "de-conversation-03-base17-block-free-u02-c01", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 85634, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 27.36, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1140480, "script_start_s": 0.16, "script_end_s": 23.76, "all_asr_last_word_s": 27.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.125, "source_identity_reward": 0.2510460001347485, "source_start_sample": 3840, "source_end_sample": 1147680, "first_word_output_sample": 3840, "last_word_output_sample": 1136640, "samples": 1143840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 3.45, "added_silence_s": 0.0, "duration_s": 23.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u02-c01.mp3" }, "de-conversation-03-base17-block-free-u02-c02": { "id": "de-conversation-03-base17-block-free-u02-c02", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 85635, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 31.36, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 54, "coverage": 0.9642857142857143, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 940800, "script_start_s": 0.16, "script_end_s": 19.6, "all_asr_last_word_s": 29.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.35714285714285715, "source_identity_reward": 0.15273206694998961, "source_start_sample": 3840, "source_end_sample": 948000, "first_word_output_sample": 3840, "last_word_output_sample": 936960, "samples": 944160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 11.61, "added_silence_s": 0.0, "duration_s": 19.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u02-c02.mp3" }, "de-conversation-03-base17-block-free-u02-c03": { "id": "de-conversation-03-base17-block-free-u02-c03", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 85636, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 24.48, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 960000, "script_start_s": 0.16, "script_end_s": 20.0, "all_asr_last_word_s": 24.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.21428571428571427, "source_identity_reward": 0.18921942067680292, "source_start_sample": 3840, "source_end_sample": 967200, "first_word_output_sample": 3840, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 4.33, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u02-c03.mp3" }, "de-conversation-03-base17-block-free-u02-c04": { "id": "de-conversation-03-base17-block-free-u02-c04", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 85637, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 25.2, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 971520, "script_start_s": 0.0, "script_end_s": 20.24, "all_asr_last_word_s": 24.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.23214285714285715, "source_identity_reward": 0.21896845543730256, "source_start_sample": 0, "source_end_sample": 978720, "first_word_output_sample": 0, "last_word_output_sample": 971520, "samples": 978720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.81, "added_silence_s": 0.0, "duration_s": 20.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u02-c04.mp3" }, "de-conversation-03-base17-block-free-u02-c05": { "id": "de-conversation-03-base17-block-free-u02-c05", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 85638, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 21.68, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 1032960, "script_start_s": 0.24, "script_end_s": 21.52, "all_asr_last_word_s": 21.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.24837610319171124, "source_start_sample": 7680, "source_end_sample": 1040160, "first_word_output_sample": 3840, "last_word_output_sample": 1025280, "samples": 1032480, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u02-c05.mp3" }, "de-conversation-03-base17-block-free-u02-c06": { "id": "de-conversation-03-base17-block-free-u02-c06", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 85639, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 28.64, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 986880, "script_start_s": 0.24, "script_end_s": 20.56, "all_asr_last_word_s": 27.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.23214285714285715, "source_identity_reward": 0.25216793393607123, "source_start_sample": 7680, "source_end_sample": 994080, "first_word_output_sample": 3840, "last_word_output_sample": 979200, "samples": 986400, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 7.93, "added_silence_s": 0.0, "duration_s": 20.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u02-c06.mp3" }, "de-conversation-03-base17-block-free-u02-c07": { "id": "de-conversation-03-base17-block-free-u02-c07", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 85640, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 31.04, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 910080, "script_start_s": 0.0, "script_end_s": 18.96, "all_asr_last_word_s": 29.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3392857142857143, "source_identity_reward": 0.24952477678346785, "source_start_sample": 0, "source_end_sample": 917280, "first_word_output_sample": 0, "last_word_output_sample": 910080, "samples": 917280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 11.93, "added_silence_s": 0.0, "duration_s": 19.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u02-c07.mp3" }, "de-conversation-03-base17-segment-timed-u01-c00": { "id": "de-conversation-03-base17-segment-timed-u01-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85117, "plain": "Und ob diese Stimmen erkennbar bleiben, wenn das Gespräch eine unerwartete Wendung nimmt?", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5566622458497891, "source_start_sample": 0, "source_end_sample": 222720, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-timed-u01-c00.mp3" }, "en-conversation-04-base17-segment-free-u01-c00": { "id": "en-conversation-04-base17-segment-free-u01-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76117, "plain": "You are kidding. Do you mean it lands on an exact millisecond?", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 15360, "last_word_sample": 264960, "script_start_s": 0.32, "script_end_s": 5.52, "all_asr_last_word_s": 5.5200000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.18904813158956787, "source_start_sample": 11520, "source_end_sample": 272160, "first_word_output_sample": 3840, "last_word_output_sample": 253440, "samples": 260640, "sample_rate": 48000, "head_removed_s": 0.24, "tail_removed_s": 24.33, "added_silence_s": 0.0, "duration_s": 5.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-free-u01-c00.mp3" }, "en-conversation-04-base17-segment-free-u11-c00": { "id": "en-conversation-04-base17-segment-free-u11-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 10.64, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 272640, "script_start_s": 0.16, "script_end_s": 5.68, "all_asr_last_word_s": 5.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.21719030142256562, "source_start_sample": 3840, "source_end_sample": 279840, "first_word_output_sample": 3840, "last_word_output_sample": 268800, "samples": 276000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 4.81, "added_silence_s": 0.0, "duration_s": 5.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-free-u11-c00.mp3" }, "de-conversation-03-base17-block-free-u01-c00": { "id": "de-conversation-03-base17-block-free-u01-c00", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 85617, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 47, "matched_tokens": 46, "coverage": 0.9787234042553191, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 952320, "script_start_s": 0.16, "script_end_s": 19.84, "all_asr_last_word_s": 23.759999999999998 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06382978723404255, "source_identity_reward": 0.3223421263639934, "source_start_sample": 3840, "source_end_sample": 959520, "first_word_output_sample": 3840, "last_word_output_sample": 948480, "samples": 955680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.01, "added_silence_s": 0.0, "duration_s": 19.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u01-c00.mp3" }, "de-conversation-03-base17-block-free-u01-c01": { "id": "de-conversation-03-base17-block-free-u01-c01", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 85618, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 27.76, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 46, "coverage": 0.9787234042553191, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 956160, "script_start_s": 0.16, "script_end_s": 19.92, "all_asr_last_word_s": 27.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2978723404255319, "source_identity_reward": 0.2684453704780333, "source_start_sample": 3840, "source_end_sample": 963360, "first_word_output_sample": 3840, "last_word_output_sample": 952320, "samples": 959520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 7.69, "added_silence_s": 0.0, "duration_s": 19.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u01-c01.mp3" }, "de-conversation-03-base17-block-free-u01-c02": { "id": "de-conversation-03-base17-block-free-u01-c02", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 85619, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 47, "matched_tokens": 45, "coverage": 0.9574468085106383, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 967680, "script_start_s": 0.0, "script_end_s": 20.16, "all_asr_last_word_s": 32.400000000000006 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06382978723404255, "source_identity_reward": 0.3158781781053378, "source_start_sample": 0, "source_end_sample": 974880, "first_word_output_sample": 0, "last_word_output_sample": 967680, "samples": 974880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 39.69, "added_silence_s": 0.0, "duration_s": 20.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u01-c02.mp3" }, "de-conversation-03-base17-block-free-u01-c03": { "id": "de-conversation-03-base17-block-free-u01-c03", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 85620, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 47, "matched_tokens": 47, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 967680, "script_start_s": 0.08, "script_end_s": 20.16, "all_asr_last_word_s": 41.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0425531914893617, "source_identity_reward": 0.2808369456415404, "source_start_sample": 0, "source_end_sample": 974880, "first_word_output_sample": 3840, "last_word_output_sample": 967680, "samples": 974880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 39.69, "added_silence_s": 0.0, "duration_s": 20.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u01-c03.mp3" }, "de-conversation-03-base17-block-free-u01-c04": { "id": "de-conversation-03-base17-block-free-u01-c04", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 85621, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 47, "matched_tokens": 45, "coverage": 0.9574468085106383, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 933120, "script_start_s": 0.16, "script_end_s": 19.44, "all_asr_last_word_s": 35.120000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1702127659574468, "source_identity_reward": 0.24962786425749, "source_start_sample": 3840, "source_end_sample": 940320, "first_word_output_sample": 3840, "last_word_output_sample": 929280, "samples": 936480, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.41, "added_silence_s": 0.0, "duration_s": 19.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u01-c04.mp3" }, "de-conversation-03-base17-block-free-u01-c05": { "id": "de-conversation-03-base17-block-free-u01-c05", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 85622, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 47, "matched_tokens": 45, "coverage": 0.9574468085106383, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1702127659574468, "source_identity_reward": 0.32793856500428653 }, "de-conversation-03-base17-block-free-u01-c06": { "id": "de-conversation-03-base17-block-free-u01-c06", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 85623, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 47, "matched_tokens": 46, "coverage": 0.9787234042553191, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 936960, "script_start_s": 0.16, "script_end_s": 19.52, "all_asr_last_word_s": 24.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0425531914893617, "source_identity_reward": 0.271649937478678, "source_start_sample": 3840, "source_end_sample": 944160, "first_word_output_sample": 3840, "last_word_output_sample": 933120, "samples": 940320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.33, "added_silence_s": 0.0, "duration_s": 19.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u01-c06.mp3" }, "de-conversation-03-base17-block-free-u01-c07": { "id": "de-conversation-03-base17-block-free-u01-c07", "case": "de-conversation-03", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 85624, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 47, "matched_tokens": 46, "coverage": 0.9787234042553191, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1006080, "script_start_s": 0.0, "script_end_s": 20.96, "all_asr_last_word_s": 59.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0851063829787234, "source_identity_reward": 0.23973395306527395, "source_start_sample": 0, "source_end_sample": 1013280, "first_word_output_sample": 0, "last_word_output_sample": 1006080, "samples": 1013280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.89, "added_silence_s": 0.0, "duration_s": 21.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-block-free-u01-c07.mp3" }, "en-conversation-02-base17-segment-free-u07-c00": { "id": "en-conversation-02-base17-segment-free-u07-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74213, "plain": "Now that is a demonstration I can enjoy. Keep the little red light on.", "source_duration_s": 7.28, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 230400, "script_start_s": 0.0, "script_end_s": 4.8, "all_asr_last_word_s": 4.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.26077007179839495, "source_start_sample": 0, "source_end_sample": 237600, "first_word_output_sample": 0, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.33, "added_silence_s": 0.0, "duration_s": 4.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-free-u07-c00.mp3" }, "de-conversation-01-base17-block-free-u00-c00": { "id": "de-conversation-01-base17-block-free-u00-c00", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 83601, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 33.52, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 40, "coverage": 0.9090909090909091, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 840960, "script_start_s": 0.0, "script_end_s": 17.52, "all_asr_last_word_s": 17.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09090909090909091, "source_identity_reward": 0.34178891007533957, "source_start_sample": 0, "source_end_sample": 848160, "first_word_output_sample": 0, "last_word_output_sample": 840960, "samples": 848160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 15.85, "added_silence_s": 0.0, "duration_s": 17.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u00-c00.mp3" }, "de-conversation-01-base17-block-free-u00-c01": { "id": "de-conversation-01-base17-block-free-u00-c01", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 83602, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 24.0, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 42, "coverage": 0.9545454545454546, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 806400, "script_start_s": 0.16, "script_end_s": 16.8, "all_asr_last_word_s": 22.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.29545454545454547, "source_identity_reward": 0.27766175538898197, "source_start_sample": 3840, "source_end_sample": 813600, "first_word_output_sample": 3840, "last_word_output_sample": 802560, "samples": 809760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 7.05, "added_silence_s": 0.0, "duration_s": 16.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u00-c01.mp3" }, "de-conversation-01-base17-block-free-u00-c02": { "id": "de-conversation-01-base17-block-free-u00-c02", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 83603, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 44, "matched_tokens": 42, "coverage": 0.9545454545454546, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 775680, "script_start_s": 0.0, "script_end_s": 16.16, "all_asr_last_word_s": 16.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06818181818181818, "source_identity_reward": 0.2628286588444942, "source_start_sample": 0, "source_end_sample": 782880, "first_word_output_sample": 0, "last_word_output_sample": 775680, "samples": 782880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 43.69, "added_silence_s": 0.0, "duration_s": 16.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u00-c02.mp3" }, "de-conversation-01-base17-block-free-u00-c03": { "id": "de-conversation-01-base17-block-free-u00-c03", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 83604, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 19.76, "source_frame_cap": false, "alignment": { "expected_tokens": 44, "matched_tokens": 40, "coverage": 0.9090909090909091, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 925440, "script_start_s": 0.0, "script_end_s": 19.28, "all_asr_last_word_s": 19.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09090909090909091, "source_identity_reward": 0.19559794641403996, "source_start_sample": 0, "source_end_sample": 932640, "first_word_output_sample": 0, "last_word_output_sample": 925440, "samples": 932640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.33, "added_silence_s": 0.0, "duration_s": 19.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u00-c03.mp3" }, "de-conversation-01-base17-block-free-u00-c04": { "id": "de-conversation-01-base17-block-free-u00-c04", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 83605, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 44, "matched_tokens": 40, "coverage": 0.9090909090909091, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 864000, "script_start_s": 0.0, "script_end_s": 18.0, "all_asr_last_word_s": 54.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5227272727272727, "source_identity_reward": 0.12822357638327353, "source_start_sample": 0, "source_end_sample": 871200, "first_word_output_sample": 0, "last_word_output_sample": 864000, "samples": 871200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 41.85, "added_silence_s": 0.0, "duration_s": 18.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u00-c04.mp3" }, "de-conversation-01-base17-block-free-u00-c05": { "id": "de-conversation-01-base17-block-free-u00-c05", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 83606, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 44, "matched_tokens": 42, "coverage": 0.9545454545454546, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 963840, "script_start_s": 0.0, "script_end_s": 20.08, "all_asr_last_word_s": 38.879999999999995 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06818181818181818, "source_identity_reward": 0.20530617394738299, "source_start_sample": 0, "source_end_sample": 971040, "first_word_output_sample": 0, "last_word_output_sample": 963840, "samples": 971040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 39.77, "added_silence_s": 0.0, "duration_s": 20.23, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u00-c05.mp3" }, "de-conversation-01-base17-block-free-u00-c06": { "id": "de-conversation-01-base17-block-free-u00-c06", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 83607, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 44, "matched_tokens": 41, "coverage": 0.9318181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 802560, "script_start_s": 0.0, "script_end_s": 16.72, "all_asr_last_word_s": 59.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5454545454545454, "source_identity_reward": 0.23101760035239335, "source_start_sample": 0, "source_end_sample": 809760, "first_word_output_sample": 0, "last_word_output_sample": 802560, "samples": 809760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 43.13, "added_silence_s": 0.0, "duration_s": 16.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u00-c06.mp3" }, "de-conversation-01-base17-block-free-u00-c07": { "id": "de-conversation-01-base17-block-free-u00-c07", "case": "de-conversation-01", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 83608, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large. Du klingst verdächtig begeistert. Hat dich ein Sprachmodell wirklich überrascht? Es liest ein Skript mit mehreren Stimmen und eigenen Spielanweisungen pro Satz. Moment, eine Stimme darf dir misstrauen, während die andere schon jubelt?", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 44, "matched_tokens": 42, "coverage": 0.9545454545454546, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 814080, "script_start_s": 0.0, "script_end_s": 16.96, "all_asr_last_word_s": 24.520000000000003 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.36363636363636365, "source_identity_reward": 0.3019054190927797, "source_start_sample": 0, "source_end_sample": 821280, "first_word_output_sample": 0, "last_word_output_sample": 814080, "samples": 821280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 42.89, "added_silence_s": 0.0, "duration_s": 17.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-block-free-u00-c07.mp3" }, "en-conversation-01-base17-segment-free-u11-c00": { "id": "en-conversation-01-base17-segment-free-u11-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 10.8, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 268800, "script_start_s": 0.16, "script_end_s": 5.6, "all_asr_last_word_s": 5.6000000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.04148624212561972, "source_start_sample": 3840, "source_end_sample": 276000, "first_word_output_sample": 3840, "last_word_output_sample": 264960, "samples": 272160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 5.05, "added_silence_s": 0.0, "duration_s": 5.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-free-u11-c00.mp3" }, "en-conversation-05-base17-segment-timed-u01-c00": { "id": "en-conversation-05-base17-segment-timed-u01-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77117, "plain": "Same model family, different voices, and a different emotional rhythm in each language?", "source_duration_s": 4.32, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 203520, "script_start_s": 0.24, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.19655030011872382, "source_start_sample": 7680, "source_end_sample": 207360, "first_word_output_sample": 3840, "last_word_output_sample": 195840, "samples": 203040, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-timed-u01-c00.mp3" }, "de-conversation-02-base17-segment-timed-u00-c00": { "id": "de-conversation-02-base17-segment-timed-u00-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84101, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 222720, "script_start_s": 0.0, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5472076909045264, "source_start_sample": 0, "source_end_sample": 226560, "first_word_output_sample": 0, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-timed-u00-c00.mp3" }, "en-conversation-05-base17-segment-free-u09-c00": { "id": "en-conversation-05-base17-segment-free-u09-c00", "case": "en-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 19.28, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 414720, "script_start_s": 0.16, "script_end_s": 8.64, "all_asr_last_word_s": 15.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-base17-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.25, "source_identity_reward": 0.32351265674388724, "source_start_sample": 3840, "source_end_sample": 421920, "first_word_output_sample": 3840, "last_word_output_sample": 410880, "samples": 418080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 10.49, "added_silence_s": 0.0, "duration_s": 8.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-base17-segment-free-u09-c00.mp3" }, "en-conversation-04-base17-segment-free-u04-c00": { "id": "en-conversation-04-base17-segment-free-u04-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76165, "plain": "We compare requested and generated durations, including early endings and frame limits.", "source_duration_s": 13.36, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 314880, "script_start_s": 0.16, "script_end_s": 6.56, "all_asr_last_word_s": 13.280000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.4166666666666667, "source_identity_reward": 0.1803066814615483, "source_start_sample": 3840, "source_end_sample": 322080, "first_word_output_sample": 3840, "last_word_output_sample": 311040, "samples": 318240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 6.65, "added_silence_s": 0.0, "duration_s": 6.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-free-u04-c00.mp3" }, "de-conversation-02-base17-segment-timed-u08-c00": { "id": "de-conversation-02-base17-segment-timed-u08-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 222720, "script_start_s": 0.0, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6007660273911279, "source_start_sample": 0, "source_end_sample": 226560, "first_word_output_sample": 0, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-timed-u08-c00.mp3" }, "en-conversation-04-base17-block-timed-u01-c00": { "id": "en-conversation-04-base17-block-timed-u01-c00", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 76617, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.96, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1002240, "script_start_s": 0.16, "script_end_s": 20.88, "all_asr_last_word_s": 20.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1718562933657581, "source_start_sample": 3840, "source_end_sample": 1006080, "first_word_output_sample": 3840, "last_word_output_sample": 998400, "samples": 1005600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u01-c00.mp3" }, "en-conversation-04-base17-block-timed-u01-c01": { "id": "en-conversation-04-base17-block-timed-u01-c01", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 76618, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 21.04, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1006080, "script_start_s": 0.16, "script_end_s": 20.96, "all_asr_last_word_s": 20.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.14181759942041763, "source_start_sample": 3840, "source_end_sample": 1009920, "first_word_output_sample": 3840, "last_word_output_sample": 1002240, "samples": 1009440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u01-c01.mp3" }, "en-conversation-04-base17-block-timed-u01-c02": { "id": "en-conversation-04-base17-block-timed-u01-c02", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 76619, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.96, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1002240, "script_start_s": 0.16, "script_end_s": 20.88, "all_asr_last_word_s": 20.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.11030050419688875, "source_start_sample": 3840, "source_end_sample": 1006080, "first_word_output_sample": 3840, "last_word_output_sample": 998400, "samples": 1005600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u01-c02.mp3" }, "en-conversation-04-base17-block-timed-u01-c03": { "id": "en-conversation-04-base17-block-timed-u01-c03", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 76620, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.96, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1002240, "script_start_s": 0.16, "script_end_s": 20.88, "all_asr_last_word_s": 20.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.21584564006902227, "source_start_sample": 3840, "source_end_sample": 1006080, "first_word_output_sample": 3840, "last_word_output_sample": 998400, "samples": 1005600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u01-c03.mp3" }, "en-conversation-04-base17-block-timed-u01-c04": { "id": "en-conversation-04-base17-block-timed-u01-c04", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 76621, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.96, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1002240, "script_start_s": 0.16, "script_end_s": 20.88, "all_asr_last_word_s": 20.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.15485454215656724, "source_start_sample": 3840, "source_end_sample": 1006080, "first_word_output_sample": 3840, "last_word_output_sample": 998400, "samples": 1005600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u01-c04.mp3" }, "en-conversation-04-base17-block-timed-u01-c05": { "id": "en-conversation-04-base17-block-timed-u01-c05", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 76622, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.96, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1002240, "script_start_s": 0.16, "script_end_s": 20.88, "all_asr_last_word_s": 20.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.14707021621998376, "source_start_sample": 3840, "source_end_sample": 1006080, "first_word_output_sample": 3840, "last_word_output_sample": 998400, "samples": 1005600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u01-c05.mp3" }, "en-conversation-04-base17-block-timed-u01-c06": { "id": "en-conversation-04-base17-block-timed-u01-c06", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 76623, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.96, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1002240, "script_start_s": 0.16, "script_end_s": 20.88, "all_asr_last_word_s": 20.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.17783697289496336, "source_start_sample": 3840, "source_end_sample": 1006080, "first_word_output_sample": 3840, "last_word_output_sample": 998400, "samples": 1005600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u01-c06.mp3" }, "en-conversation-04-base17-block-timed-u01-c07": { "id": "en-conversation-04-base17-block-timed-u01-c07", "case": "en-conversation-04", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 76624, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 20.96, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1002240, "script_start_s": 0.16, "script_end_s": 20.88, "all_asr_last_word_s": 20.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2009845029126665, "source_start_sample": 3840, "source_end_sample": 1006080, "first_word_output_sample": 3840, "last_word_output_sample": 998400, "samples": 1005600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-block-timed-u01-c07.mp3" }, "de-conversation-03-base17-segment-timed-u10-c00": { "id": "de-conversation-03-base17-segment-timed-u10-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 5.68, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 268800, "script_start_s": 0.16, "script_end_s": 5.6, "all_asr_last_word_s": 5.6000000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6026552260252567, "source_start_sample": 3840, "source_end_sample": 272640, "first_word_output_sample": 3840, "last_word_output_sample": 264960, "samples": 272160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-timed-u10-c00.mp3" }, "de-conversation-05-base17-segment-timed-u02-c00": { "id": "de-conversation-05-base17-segment-timed-u02-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87133, "plain": "Das untersucht dieses Hörgrid, mit kurzen Dialogen und längeren Gesprächen.", "source_duration_s": 3.68, "source_frame_cap": false, "alignment": { "expected_tokens": 10, "matched_tokens": 9, "coverage": 0.9, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 172800, "script_start_s": 0.0, "script_end_s": 3.6, "all_asr_last_word_s": 3.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1, "source_identity_reward": 0.7473147127240223, "source_start_sample": 0, "source_end_sample": 176640, "first_word_output_sample": 0, "last_word_output_sample": 172800, "samples": 180000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-timed-u02-c00.mp3" }, "en-conversation-03-base17-block-timed-u02-c00": { "id": "en-conversation-03-base17-block-timed-u02-c00", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 75633, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1214178840162527, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u02-c00.mp3" }, "en-conversation-03-base17-block-timed-u02-c01": { "id": "en-conversation-03-base17-block-timed-u02-c01", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 75634, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 879360, "script_start_s": 0.24, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1519992266876667, "source_start_sample": 7680, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u02-c01.mp3" }, "en-conversation-03-base17-block-timed-u02-c02": { "id": "en-conversation-03-base17-block-timed-u02-c02", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 75635, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.15575690543028264, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u02-c02.mp3" }, "en-conversation-03-base17-block-timed-u02-c03": { "id": "en-conversation-03-base17-block-timed-u02-c03", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 75636, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.12247737323789451, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u02-c03.mp3" }, "en-conversation-03-base17-block-timed-u02-c04": { "id": "en-conversation-03-base17-block-timed-u02-c04", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 75637, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.10005979961067542, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u02-c04.mp3" }, "en-conversation-03-base17-block-timed-u02-c05": { "id": "en-conversation-03-base17-block-timed-u02-c05", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 75638, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.062157052147755174, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u02-c05.mp3" }, "en-conversation-03-base17-block-timed-u02-c06": { "id": "en-conversation-03-base17-block-timed-u02-c06", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 75639, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.07233950806604228, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u02-c06.mp3" }, "en-conversation-03-base17-block-timed-u02-c07": { "id": "en-conversation-03-base17-block-timed-u02-c07", "case": "en-conversation-03", "model": "base17", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 75640, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1214127518941249, "source_start_sample": 3840, "source_end_sample": 883200, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-block-timed-u02-c07.mp3" }, "en-conversation-01-base17-segment-timed-u02-c00": { "id": "en-conversation-01-base17-segment-timed-u02-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73133, "plain": "It can read a script with different speakers and sentence-level acting directions.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 192000, "script_start_s": 0.0, "script_end_s": 4.0, "all_asr_last_word_s": 4.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4803216789134861, "source_start_sample": 0, "source_end_sample": 195840, "first_word_output_sample": 0, "last_word_output_sample": 192000, "samples": 199200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-timed-u02-c00.mp3" }, "de-conversation-04-base17-segment-timed-u05-c00": { "id": "de-conversation-04-base17-segment-timed-u05-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86181, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen.", "source_duration_s": 5.36, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 249600, "script_start_s": 0.08, "script_end_s": 5.2, "all_asr_last_word_s": 5.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.48191390402576734, "source_start_sample": 0, "source_end_sample": 256800, "first_word_output_sample": 3840, "last_word_output_sample": 249600, "samples": 256800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-timed-u05-c00.mp3" }, "de-conversation-01-base17-segment-timed-u08-c00": { "id": "de-conversation-01-base17-segment-timed-u08-c00", "case": "de-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-base17-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6160534596914431, "source_start_sample": 0, "source_end_sample": 222720, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-base17-segment-timed-u08-c00.mp3" }, "en-conversation-04-base17-segment-free-u07-c00": { "id": "en-conversation-04-base17-segment-free-u07-c00", "case": "en-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76213, "plain": "A reassuringly human approach to a very peculiar little machine.", "source_duration_s": 9.84, "source_frame_cap": false, "alignment": { "expected_tokens": 10, "matched_tokens": 10, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 468480, "script_start_s": 0.24, "script_end_s": 9.76, "all_asr_last_word_s": 9.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-base17-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.045852635027852286, "source_start_sample": 7680, "source_end_sample": 472320, "first_word_output_sample": 3840, "last_word_output_sample": 460800, "samples": 468000, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 9.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-base17-segment-free-u07-c00.mp3" }, "de-conversation-02-base17-segment-free-u04-c00": { "id": "de-conversation-02-base17-segment-free-u04-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84165, "plain": "Unbedingt. Jede Äußerung bekommt ihren eigenen Text und eine eigene emotionale Richtung.", "source_duration_s": 18.64, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 322560, "script_start_s": 0.16, "script_end_s": 6.72, "all_asr_last_word_s": 15.120000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5833333333333334, "source_identity_reward": 0.4432769076374186, "source_start_sample": 3840, "source_end_sample": 329760, "first_word_output_sample": 3840, "last_word_output_sample": 318720, "samples": 325920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 11.77, "added_silence_s": 0.0, "duration_s": 6.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-free-u04-c00.mp3" }, "en-conversation-03-base17-segment-free-u00-c00": { "id": "en-conversation-03-base17-segment-free-u00-c00", "case": "en-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75101, "plain": "Here is the part that interests me: using a short recording to establish each voice.", "source_duration_s": 9.6, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 288000, "script_start_s": 0.0, "script_end_s": 6.0, "all_asr_last_word_s": 6.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-base17-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4367226408704159, "source_start_sample": 0, "source_end_sample": 295200, "first_word_output_sample": 0, "last_word_output_sample": 288000, "samples": 295200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 3.45, "added_silence_s": 0.0, "duration_s": 6.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-base17-segment-free-u00-c00.mp3" }, "de-conversation-05-base17-segment-timed-u13-c00": { "id": "de-conversation-05-base17-segment-timed-u13-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87309, "plain": "Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden.", "source_duration_s": 7.12, "source_frame_cap": false, "alignment": { "expected_tokens": 20, "matched_tokens": 18, "coverage": 0.9, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 15360, "last_word_sample": 334080, "script_start_s": 0.32, "script_end_s": 6.96, "all_asr_last_word_s": 6.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-timed-u13-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1, "source_identity_reward": 0.6130817824315115, "source_start_sample": 11520, "source_end_sample": 341280, "first_word_output_sample": 3840, "last_word_output_sample": 322560, "samples": 329760, "sample_rate": 48000, "head_removed_s": 0.24, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 6.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-timed-u13-c00.mp3" }, "de-conversation-03-base17-segment-free-u10-c00": { "id": "de-conversation-03-base17-segment-free-u10-c00", "case": "de-conversation-03", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 629760, "script_start_s": 0.24, "script_end_s": 13.12, "all_asr_last_word_s": 29.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-base17-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5625, "source_identity_reward": 0.4491995734238284, "source_start_sample": 7680, "source_end_sample": 636960, "first_word_output_sample": 3840, "last_word_output_sample": 622080, "samples": 629280, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 16.73, "added_silence_s": 0.0, "duration_s": 13.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-base17-segment-free-u10-c00.mp3" }, "de-conversation-02-base17-segment-free-u10-c00": { "id": "de-conversation-02-base17-segment-free-u10-c00", "case": "de-conversation-02", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 368640, "script_start_s": 0.16, "script_end_s": 7.68, "all_asr_last_word_s": 14.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-base17-segment-free-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4375, "source_identity_reward": 0.5535368122578608, "source_start_sample": 3840, "source_end_sample": 375840, "first_word_output_sample": 3840, "last_word_output_sample": 364800, "samples": 372000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 22.17, "added_silence_s": 0.0, "duration_s": 7.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-base17-segment-free-u10-c00.mp3" }, "de-conversation-05-base17-segment-free-u13-c00": { "id": "de-conversation-05-base17-segment-free-u13-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87309, "plain": "Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 20, "matched_tokens": 20, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 384000, "script_start_s": 0.16, "script_end_s": 8.0, "all_asr_last_word_s": 29.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u13-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.65, "source_identity_reward": 0.35301671234964244, "source_start_sample": 3840, "source_end_sample": 391200, "first_word_output_sample": 3840, "last_word_output_sample": 380160, "samples": 387360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 21.85, "added_silence_s": 0.0, "duration_s": 8.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u13-c00.mp3" }, "en-conversation-01-base17-segment-timed-u10-c00": { "id": "en-conversation-01-base17-segment-timed-u10-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 4.64, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 215040, "script_start_s": 0.0, "script_end_s": 4.48, "all_asr_last_word_s": 4.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4053874724782013, "source_start_sample": 0, "source_end_sample": 222240, "first_word_output_sample": 0, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.63, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-base17-segment-timed-u10-c00.mp3" }, "en-conversation-02-base17-segment-timed-u09-c00": { "id": "en-conversation-02-base17-segment-timed-u09-c00", "case": "en-conversation-02", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 4.0, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 188160, "script_start_s": 0.16, "script_end_s": 3.92, "all_asr_last_word_s": 3.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.0981552459065446, "source_start_sample": 3840, "source_end_sample": 192000, "first_word_output_sample": 3840, "last_word_output_sample": 184320, "samples": 191520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 3.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-segment-timed-u09-c00.mp3" }, "en-conversation-02-base17-block-free-u02-c00": { "id": "en-conversation-02-base17-block-free-u02-c00", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 74633, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 537600, "script_start_s": 0.16, "script_end_s": 11.2, "all_asr_last_word_s": 41.400000000000006 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4482758620689655, "source_identity_reward": 0.08847241996290967, "source_start_sample": 3840, "source_end_sample": 544800, "first_word_output_sample": 3840, "last_word_output_sample": 533760, "samples": 540960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 48.65, "added_silence_s": 0.0, "duration_s": 11.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u02-c00.mp3" }, "en-conversation-02-base17-block-free-u02-c01": { "id": "en-conversation-02-base17-block-free-u02-c01", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 74634, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 56.0, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 487680, "script_start_s": 0.0, "script_end_s": 10.16, "all_asr_last_word_s": 51.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4827586206896552, "source_identity_reward": 0.152222137255442, "source_start_sample": 0, "source_end_sample": 494880, "first_word_output_sample": 0, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 45.69, "added_silence_s": 0.0, "duration_s": 10.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u02-c01.mp3" }, "en-conversation-02-base17-block-free-u02-c02": { "id": "en-conversation-02-base17-block-free-u02-c02", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 74635, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 10.64, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 499200, "script_start_s": 0.16, "script_end_s": 10.4, "all_asr_last_word_s": 10.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.13088128094352966, "source_start_sample": 3840, "source_end_sample": 506400, "first_word_output_sample": 3840, "last_word_output_sample": 495360, "samples": 502560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 10.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u02-c02.mp3" }, "en-conversation-02-base17-block-free-u02-c03": { "id": "en-conversation-02-base17-block-free-u02-c03", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 74636, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 520320, "script_start_s": 0.0, "script_end_s": 10.84, "all_asr_last_word_s": 59.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.8275862068965517, "source_identity_reward": 0.08555206195624179, "source_start_sample": 0, "source_end_sample": 527520, "first_word_output_sample": 0, "last_word_output_sample": 520320, "samples": 527520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 49.01, "added_silence_s": 0.0, "duration_s": 10.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u02-c03.mp3" }, "en-conversation-02-base17-block-free-u02-c04": { "id": "en-conversation-02-base17-block-free-u02-c04", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 74637, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 25.84, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 476160, "script_start_s": 0.0, "script_end_s": 9.92, "all_asr_last_word_s": 9.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.10948404879974907, "source_start_sample": 0, "source_end_sample": 483360, "first_word_output_sample": 0, "last_word_output_sample": 476160, "samples": 483360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 15.77, "added_silence_s": 0.0, "duration_s": 10.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u02-c04.mp3" }, "en-conversation-02-base17-block-free-u02-c05": { "id": "en-conversation-02-base17-block-free-u02-c05", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 74638, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 14.16, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 510720, "script_start_s": 0.0, "script_end_s": 10.64, "all_asr_last_word_s": 10.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1267647117703681, "source_start_sample": 0, "source_end_sample": 517920, "first_word_output_sample": 0, "last_word_output_sample": 510720, "samples": 517920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 3.37, "added_silence_s": 0.0, "duration_s": 10.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u02-c05.mp3" }, "en-conversation-02-base17-block-free-u02-c06": { "id": "en-conversation-02-base17-block-free-u02-c06", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 74639, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 12.72, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 476160, "script_start_s": 0.16, "script_end_s": 9.92, "all_asr_last_word_s": 9.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.18781968419448047, "source_start_sample": 3840, "source_end_sample": 483360, "first_word_output_sample": 3840, "last_word_output_sample": 472320, "samples": 479520, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 2.65, "added_silence_s": 0.0, "duration_s": 9.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u02-c06.mp3" }, "en-conversation-02-base17-block-free-u02-c07": { "id": "en-conversation-02-base17-block-free-u02-c07", "case": "en-conversation-02", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 74640, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 518400, "script_start_s": 0.0, "script_end_s": 10.8, "all_asr_last_word_s": 23.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-base17-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.10344827586206896, "source_identity_reward": 0.11261864746523303, "source_start_sample": 0, "source_end_sample": 525600, "first_word_output_sample": 0, "last_word_output_sample": 518400, "samples": 525600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 49.05, "added_silence_s": 0.0, "duration_s": 10.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-base17-block-free-u02-c07.mp3" }, "en-conversation-01-base17-segment-free-u07-c00": { "id": "en-conversation-01-base17-segment-free-u07-c00", "case": "en-conversation-01", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73213, "plain": "All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 11.2, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 12, "coverage": 0.8571428571428571, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-base17-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.14285714285714285, "source_identity_reward": 0.23578133419758462 }, "de-conversation-04-base17-segment-timed-u10-c00": { "id": "de-conversation-04-base17-segment-timed-u10-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86261, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht.", "source_duration_s": 5.68, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 264960, "script_start_s": 0.08, "script_end_s": 5.52, "all_asr_last_word_s": 5.5200000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7171367022239721, "source_start_sample": 0, "source_end_sample": 272160, "first_word_output_sample": 3840, "last_word_output_sample": 264960, "samples": 272160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-timed-u10-c00.mp3" }, "de-conversation-04-base17-segment-free-u07-c00": { "id": "de-conversation-04-base17-segment-free-u07-c00", "case": "de-conversation-04", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86213, "plain": "Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 10, "matched_tokens": 10, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 15360, "last_word_sample": 303360, "script_start_s": 0.32, "script_end_s": 6.32, "all_asr_last_word_s": 6.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-base17-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7227146320369264, "source_start_sample": 11520, "source_end_sample": 310560, "first_word_output_sample": 3840, "last_word_output_sample": 291840, "samples": 299040, "sample_rate": 48000, "head_removed_s": 0.24, "tail_removed_s": 23.53, "added_silence_s": 0.0, "duration_s": 6.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-base17-segment-free-u07-c00.mp3" }, "de-conversation-05-base17-segment-free-u14-c00": { "id": "de-conversation-05-base17-segment-free-u14-c00", "case": "de-conversation-05", "model": "base17", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87325, "plain": "Genau. Das Gespräch soll aufeinander reagieren, und nach jedem Stimmungswechsel sollen dieselben zwei Menschen wiederzuerkennen sein.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 16, "matched_tokens": 15, "coverage": 0.9375, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 360960, "script_start_s": 0.16, "script_end_s": 7.52, "all_asr_last_word_s": 7.5200000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-segment-free-u14-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.1875, "source_identity_reward": 0.6155595547728507, "source_start_sample": 3840, "source_end_sample": 368160, "first_word_output_sample": 3840, "last_word_output_sample": 357120, "samples": 364320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 22.33, "added_silence_s": 0.0, "duration_s": 7.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-segment-free-u14-c00.mp3" }, "de-conversation-05-base17-block-free-u01-c00": { "id": "de-conversation-05-base17-block-free-u01-c00", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 0, "seed": 87617, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 20.24, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 53, "coverage": 0.9814814814814815, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 963840, "script_start_s": 0.16, "script_end_s": 20.08, "all_asr_last_word_s": 20.080000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.018518518518518517, "source_identity_reward": 0.40823429708862186, "source_start_sample": 3840, "source_end_sample": 971040, "first_word_output_sample": 3840, "last_word_output_sample": 960000, "samples": 967200, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u01-c00.mp3" }, "de-conversation-05-base17-block-free-u01-c01": { "id": "de-conversation-05-base17-block-free-u01-c01", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 1, "seed": 87618, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 977280, "script_start_s": 0.08, "script_end_s": 20.36, "all_asr_last_word_s": 20.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05555555555555555, "source_identity_reward": 0.2857784306949602, "source_start_sample": 0, "source_end_sample": 984480, "first_word_output_sample": 3840, "last_word_output_sample": 977280, "samples": 984480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 39.49, "added_silence_s": 0.0, "duration_s": 20.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u01-c01.mp3" }, "de-conversation-05-base17-block-free-u01-c02": { "id": "de-conversation-05-base17-block-free-u01-c02", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 2, "seed": 87619, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 54, "matched_tokens": 53, "coverage": 0.9814814814814815, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 948480, "script_start_s": 0.16, "script_end_s": 19.76, "all_asr_last_word_s": 25.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2777777777777778, "source_identity_reward": 0.1812351222231261, "source_start_sample": 3840, "source_end_sample": 955680, "first_word_output_sample": 3840, "last_word_output_sample": 944640, "samples": 951840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.09, "added_silence_s": 0.0, "duration_s": 19.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u01-c02.mp3" }, "de-conversation-05-base17-block-free-u01-c03": { "id": "de-conversation-05-base17-block-free-u01-c03", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 3, "seed": 87620, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 938880, "script_start_s": 0.16, "script_end_s": 19.56, "all_asr_last_word_s": 19.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3870076541196191, "source_start_sample": 3840, "source_end_sample": 946080, "first_word_output_sample": 3840, "last_word_output_sample": 935040, "samples": 942240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.29, "added_silence_s": 0.0, "duration_s": 19.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u01-c03.mp3" }, "de-conversation-05-base17-block-free-u01-c04": { "id": "de-conversation-05-base17-block-free-u01-c04", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 4, "seed": 87621, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 28.4, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 51, "coverage": 0.9444444444444444, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 890880, "script_start_s": 0.16, "script_end_s": 18.56, "all_asr_last_word_s": 28.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.37037037037037035, "source_identity_reward": 0.17730745853456772, "source_start_sample": 3840, "source_end_sample": 898080, "first_word_output_sample": 3840, "last_word_output_sample": 887040, "samples": 894240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 9.69, "added_silence_s": 0.0, "duration_s": 18.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u01-c04.mp3" }, "de-conversation-05-base17-block-free-u01-c05": { "id": "de-conversation-05-base17-block-free-u01-c05", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 5, "seed": 87622, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 54, "matched_tokens": 53, "coverage": 0.9814814814814815, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 944640, "script_start_s": 0.16, "script_end_s": 19.68, "all_asr_last_word_s": 25.599999999999998 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2777777777777778, "source_identity_reward": 0.30338788385423804, "source_start_sample": 3840, "source_end_sample": 951840, "first_word_output_sample": 3840, "last_word_output_sample": 940800, "samples": 948000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.17, "added_silence_s": 0.0, "duration_s": 19.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u01-c05.mp3" }, "de-conversation-05-base17-block-free-u01-c06": { "id": "de-conversation-05-base17-block-free-u01-c06", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 6, "seed": 87623, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 19.84, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 944640, "script_start_s": 0.16, "script_end_s": 19.68, "all_asr_last_word_s": 19.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3627177370628595, "source_start_sample": 3840, "source_end_sample": 951840, "first_word_output_sample": 3840, "last_word_output_sample": 940800, "samples": 948000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u01-c06.mp3" }, "de-conversation-05-base17-block-free-u01-c07": { "id": "de-conversation-05-base17-block-free-u01-c07", "case": "de-conversation-05", "model": "base17", "strategy": "block", "timing": "free", "candidate": 7, "seed": 87624, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 994560, "script_start_s": 0.16, "script_end_s": 20.72, "all_asr_last_word_s": 20.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-base17-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.29640716149962654, "source_start_sample": 3840, "source_end_sample": 1001760, "first_word_output_sample": 3840, "last_word_output_sample": 990720, "samples": 997920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 39.13, "added_silence_s": 0.0, "duration_s": 20.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-base17-block-free-u01-c07.mp3" }, "de-conversation-01-podcast-segment-free-u09-c00": { "id": "de-conversation-01-podcast-segment-free-u09-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 83245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 9.44, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 399360, "script_start_s": 0.0, "script_end_s": 8.32, "all_asr_last_word_s": 9.200000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.21428571428571427, "source_identity_reward": 0.3820749214267512, "source_start_sample": 0, "source_end_sample": 406560, "first_word_output_sample": 0, "last_word_output_sample": 399360, "samples": 406560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.97, "added_silence_s": 0.0, "duration_s": 8.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-free-u09-c00.mp3" }, "en-conversation-01-podcast-segment-timed-u00-c00": { "id": "en-conversation-01-podcast-segment-timed-u00-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73101, "plain": "I have been listening to Humaneness Podcast, a fine-tune of Humaneness Voice Large.", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 10, "coverage": 0.7142857142857143, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2857142857142857, "source_identity_reward": 0.3469628100148764 }, "de-conversation-05-podcast-segment-free-u15-c00": { "id": "de-conversation-05-podcast-segment-free-u15-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87341, "plain": "Dann ist es ein fairer Vergleich. Wir lassen die Worte gleich und hören, was die verschiedenen Verfahren daraus machen.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 19, "matched_tokens": 19, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 322560, "script_start_s": 0.0, "script_end_s": 6.72, "all_asr_last_word_s": 6.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u15-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4691261862067149, "source_start_sample": 0, "source_end_sample": 329760, "first_word_output_sample": 0, "last_word_output_sample": 322560, "samples": 329760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.13, "added_silence_s": 0.0, "duration_s": 6.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u15-c00.mp3" }, "en-conversation-04-podcast-segment-timed-u05-c00": { "id": "en-conversation-04-podcast-segment-timed-u05-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76181, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 222720, "script_start_s": 0.16, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.307567344739763, "source_start_sample": 3840, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-timed-u05-c00.mp3" }, "de-conversation-03-podcast-segment-free-u04-c00": { "id": "de-conversation-03-podcast-segment-free-u04-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85165, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 257280, "script_start_s": 0.08, "script_end_s": 5.36, "all_asr_last_word_s": 29.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0, "source_identity_reward": 0.4276278476247847, "source_start_sample": 0, "source_end_sample": 264480, "first_word_output_sample": 3840, "last_word_output_sample": 257280, "samples": 264480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.49, "added_silence_s": 0.0, "duration_s": 5.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-free-u04-c00.mp3" }, "de-conversation-02-podcast-block-free-u02-c00": { "id": "de-conversation-02-podcast-block-free-u02-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 84633, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 11.28, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 27, "coverage": 0.9310344827586207, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 533760, "script_start_s": 0.0, "script_end_s": 11.12, "all_asr_last_word_s": 11.120000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06896551724137931, "source_identity_reward": 0.6818511958633783, "source_start_sample": 0, "source_end_sample": 540960, "first_word_output_sample": 0, "last_word_output_sample": 533760, "samples": 540960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 11.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u02-c00.mp3" }, "de-conversation-02-podcast-block-free-u02-c01": { "id": "de-conversation-02-podcast-block-free-u02-c01", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 84634, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 11.28, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 27, "coverage": 0.9310344827586207, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 533760, "script_start_s": 0.0, "script_end_s": 11.12, "all_asr_last_word_s": 11.120000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06896551724137931, "source_identity_reward": 0.7540502474760684, "source_start_sample": 0, "source_end_sample": 540960, "first_word_output_sample": 0, "last_word_output_sample": 533760, "samples": 540960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 11.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u02-c01.mp3" }, "de-conversation-02-podcast-block-free-u02-c02": { "id": "de-conversation-02-podcast-block-free-u02-c02", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 84635, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 526080, "script_start_s": 0.0, "script_end_s": 10.96, "all_asr_last_word_s": 25.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4482758620689655, "source_identity_reward": 0.5516053470880272, "source_start_sample": 0, "source_end_sample": 533280, "first_word_output_sample": 0, "last_word_output_sample": 526080, "samples": 533280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 48.89, "added_silence_s": 0.0, "duration_s": 11.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u02-c02.mp3" }, "de-conversation-02-podcast-block-free-u02-c03": { "id": "de-conversation-02-podcast-block-free-u02-c03", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 84636, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 12.08, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 533760, "script_start_s": 0.0, "script_end_s": 11.12, "all_asr_last_word_s": 11.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.034482758620689655, "source_identity_reward": 0.7935607128497233, "source_start_sample": 0, "source_end_sample": 540960, "first_word_output_sample": 0, "last_word_output_sample": 533760, "samples": 540960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.81, "added_silence_s": 0.0, "duration_s": 11.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u02-c03.mp3" }, "de-conversation-02-podcast-block-free-u02-c04": { "id": "de-conversation-02-podcast-block-free-u02-c04", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 84637, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 541440, "script_start_s": 0.0, "script_end_s": 11.28, "all_asr_last_word_s": 11.280000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7941103893986581, "source_start_sample": 0, "source_end_sample": 548640, "first_word_output_sample": 0, "last_word_output_sample": 541440, "samples": 548640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 48.57, "added_silence_s": 0.0, "duration_s": 11.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u02-c04.mp3" }, "de-conversation-02-podcast-block-free-u02-c05": { "id": "de-conversation-02-podcast-block-free-u02-c05", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 84638, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 568320, "script_start_s": 0.16, "script_end_s": 11.84, "all_asr_last_word_s": 59.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 2.586206896551724, "source_identity_reward": 0.21792928596235633, "source_start_sample": 3840, "source_end_sample": 575520, "first_word_output_sample": 3840, "last_word_output_sample": 564480, "samples": 571680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 48.01, "added_silence_s": 0.0, "duration_s": 11.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u02-c05.mp3" }, "de-conversation-02-podcast-block-free-u02-c06": { "id": "de-conversation-02-podcast-block-free-u02-c06", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 84639, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 11.04, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 522240, "script_start_s": 0.0, "script_end_s": 10.88, "all_asr_last_word_s": 10.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8713014898301291, "source_start_sample": 0, "source_end_sample": 529440, "first_word_output_sample": 0, "last_word_output_sample": 522240, "samples": 529440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 11.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u02-c06.mp3" }, "de-conversation-02-podcast-block-free-u02-c07": { "id": "de-conversation-02-podcast-block-free-u02-c07", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 84640, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.0, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 514560, "script_start_s": 0.08, "script_end_s": 10.72, "all_asr_last_word_s": 19.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.6896551724137931, "source_identity_reward": 0.4787573457801224, "source_start_sample": 0, "source_end_sample": 521760, "first_word_output_sample": 3840, "last_word_output_sample": 514560, "samples": 521760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 9.13, "added_silence_s": 0.0, "duration_s": 10.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-free-u02-c07.mp3" }, "en-conversation-03-podcast-segment-timed-u10-c00": { "id": "en-conversation-03-podcast-segment-timed-u10-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75261, "plain": "So let us compare the voices carefully, including the moments where the illusion breaks.", "source_duration_s": 4.8, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 226560, "script_start_s": 0.0, "script_end_s": 4.72, "all_asr_last_word_s": 4.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-timed-u10-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.1288676741499191, "source_start_sample": 0, "source_end_sample": 230400, "first_word_output_sample": 0, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-timed-u10-c00.mp3" }, "de-conversation-04-podcast-block-free-u02-c00": { "id": "de-conversation-04-podcast-block-free-u02-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 86633, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 11.52, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 545280, "script_start_s": 0.0, "script_end_s": 11.36, "all_asr_last_word_s": 11.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8173735002722942, "source_start_sample": 0, "source_end_sample": 552480, "first_word_output_sample": 0, "last_word_output_sample": 545280, "samples": 552480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 11.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u02-c00.mp3" }, "de-conversation-04-podcast-block-free-u02-c01": { "id": "de-conversation-04-podcast-block-free-u02-c01", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 86634, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 11.2, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 529920, "script_start_s": 0.0, "script_end_s": 11.04, "all_asr_last_word_s": 11.040000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.808351379991285, "source_start_sample": 0, "source_end_sample": 537120, "first_word_output_sample": 0, "last_word_output_sample": 529920, "samples": 537120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 11.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u02-c01.mp3" }, "de-conversation-04-podcast-block-free-u02-c02": { "id": "de-conversation-04-podcast-block-free-u02-c02", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 86635, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 19.92, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 537600, "script_start_s": 0.0, "script_end_s": 11.2, "all_asr_last_word_s": 19.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5172413793103449, "source_identity_reward": 0.514456610607971, "source_start_sample": 0, "source_end_sample": 544800, "first_word_output_sample": 0, "last_word_output_sample": 537600, "samples": 544800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 8.57, "added_silence_s": 0.0, "duration_s": 11.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u02-c02.mp3" }, "de-conversation-04-podcast-block-free-u02-c03": { "id": "de-conversation-04-podcast-block-free-u02-c03", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 86636, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 11.2, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 529920, "script_start_s": 0.0, "script_end_s": 11.04, "all_asr_last_word_s": 11.040000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7559155714193564, "source_start_sample": 0, "source_end_sample": 537120, "first_word_output_sample": 0, "last_word_output_sample": 529920, "samples": 537120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 11.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u02-c03.mp3" }, "de-conversation-04-podcast-block-free-u02-c04": { "id": "de-conversation-04-podcast-block-free-u02-c04", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 86637, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 16.4, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 560640, "script_start_s": 0.0, "script_end_s": 11.68, "all_asr_last_word_s": 15.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.13793103448275862, "source_identity_reward": 0.6292965762917392, "source_start_sample": 0, "source_end_sample": 567840, "first_word_output_sample": 0, "last_word_output_sample": 560640, "samples": 567840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.57, "added_silence_s": 0.0, "duration_s": 11.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u02-c04.mp3" }, "de-conversation-04-podcast-block-free-u02-c05": { "id": "de-conversation-04-podcast-block-free-u02-c05", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 86638, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 11.6, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 549120, "script_start_s": 0.0, "script_end_s": 11.44, "all_asr_last_word_s": 11.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8165651403017269, "source_start_sample": 0, "source_end_sample": 556320, "first_word_output_sample": 0, "last_word_output_sample": 549120, "samples": 556320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 11.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u02-c05.mp3" }, "de-conversation-04-podcast-block-free-u02-c06": { "id": "de-conversation-04-podcast-block-free-u02-c06", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 86639, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 19.84, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 518400, "script_start_s": 0.0, "script_end_s": 10.8, "all_asr_last_word_s": 19.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4482758620689655, "source_identity_reward": 0.4709125000841641, "source_start_sample": 0, "source_end_sample": 525600, "first_word_output_sample": 0, "last_word_output_sample": 518400, "samples": 525600, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 8.89, "added_silence_s": 0.0, "duration_s": 10.95, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u02-c06.mp3" }, "de-conversation-04-podcast-block-free-u02-c07": { "id": "de-conversation-04-podcast-block-free-u02-c07", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 86640, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 549120, "script_start_s": 0.0, "script_end_s": 11.44, "all_asr_last_word_s": 59.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 2.6206896551724137, "source_identity_reward": 0.22229316459367904, "source_start_sample": 0, "source_end_sample": 556320, "first_word_output_sample": 0, "last_word_output_sample": 549120, "samples": 556320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 48.41, "added_silence_s": 0.0, "duration_s": 11.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-free-u02-c07.mp3" }, "de-conversation-02-podcast-segment-timed-u02-c00": { "id": "de-conversation-02-podcast-segment-timed-u02-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84133, "plain": "Natürlich. Humaneness Voice Large liefert die Grundlage, das Podcasttraining übt den Dialog.", "source_duration_s": 4.48, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 8, "coverage": 0.6666666666666666, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 211200, "script_start_s": 0.16, "script_end_s": 4.4, "all_asr_last_word_s": 4.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.4166666666666667, "source_identity_reward": 0.4428444155853839, "source_start_sample": 3840, "source_end_sample": 215040, "first_word_output_sample": 3840, "last_word_output_sample": 207360, "samples": 214560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-timed-u02-c00.mp3" }, "de-conversation-05-podcast-segment-free-u06-c00": { "id": "de-conversation-05-podcast-segment-free-u06-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87197, "plain": "Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 299520, "script_start_s": 0.08, "script_end_s": 6.24, "all_asr_last_word_s": 10.08 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5384615384615384, "source_identity_reward": 0.4197661675325224, "source_start_sample": 0, "source_end_sample": 306720, "first_word_output_sample": 3840, "last_word_output_sample": 299520, "samples": 306720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 23.61, "added_silence_s": 0.0, "duration_s": 6.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u06-c00.mp3" }, "de-conversation-05-podcast-block-free-u02-c00": { "id": "de-conversation-05-podcast-block-free-u02-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 87633, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 21.44, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1021440, "script_start_s": 0.0, "script_end_s": 21.28, "all_asr_last_word_s": 21.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7857932392303665, "source_start_sample": 0, "source_end_sample": 1028640, "first_word_output_sample": 0, "last_word_output_sample": 1021440, "samples": 1028640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u02-c00.mp3" }, "de-conversation-05-podcast-block-free-u02-c01": { "id": "de-conversation-05-podcast-block-free-u02-c01", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 87634, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 25.6, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1029120, "script_start_s": 0.0, "script_end_s": 21.44, "all_asr_last_word_s": 21.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.7502447161693648, "source_start_sample": 0, "source_end_sample": 1036320, "first_word_output_sample": 0, "last_word_output_sample": 1029120, "samples": 1036320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.01, "added_silence_s": 0.0, "duration_s": 21.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u02-c01.mp3" }, "de-conversation-05-podcast-block-free-u02-c02": { "id": "de-conversation-05-podcast-block-free-u02-c02", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 87635, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 21.28, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1013760, "script_start_s": 0.0, "script_end_s": 21.12, "all_asr_last_word_s": 21.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.7861516657873252, "source_start_sample": 0, "source_end_sample": 1020960, "first_word_output_sample": 0, "last_word_output_sample": 1013760, "samples": 1020960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u02-c02.mp3" }, "de-conversation-05-podcast-block-free-u02-c03": { "id": "de-conversation-05-podcast-block-free-u02-c03", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 87636, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.08, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.7698699853575777, "source_start_sample": 0, "source_end_sample": 963360, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u02-c03.mp3" }, "de-conversation-05-podcast-block-free-u02-c04": { "id": "de-conversation-05-podcast-block-free-u02-c04", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 87637, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 21.36, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1017600, "script_start_s": 0.0, "script_end_s": 21.2, "all_asr_last_word_s": 21.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.7729590258710981, "source_start_sample": 0, "source_end_sample": 1024800, "first_word_output_sample": 0, "last_word_output_sample": 1017600, "samples": 1024800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u02-c04.mp3" }, "de-conversation-05-podcast-block-free-u02-c05": { "id": "de-conversation-05-podcast-block-free-u02-c05", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 87638, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 28.24, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1006080, "script_start_s": 0.0, "script_end_s": 20.96, "all_asr_last_word_s": 28.080000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.21428571428571427, "source_identity_reward": 0.6567516411351466, "source_start_sample": 0, "source_end_sample": 1013280, "first_word_output_sample": 0, "last_word_output_sample": 1006080, "samples": 1013280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 7.13, "added_silence_s": 0.0, "duration_s": 21.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u02-c05.mp3" }, "de-conversation-05-podcast-block-free-u02-c06": { "id": "de-conversation-05-podcast-block-free-u02-c06", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 87639, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.56, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 979200, "script_start_s": 0.0, "script_end_s": 20.4, "all_asr_last_word_s": 20.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03571428571428571, "source_identity_reward": 0.8053292638398136, "source_start_sample": 0, "source_end_sample": 986400, "first_word_output_sample": 0, "last_word_output_sample": 979200, "samples": 986400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u02-c06.mp3" }, "de-conversation-05-podcast-block-free-u02-c07": { "id": "de-conversation-05-podcast-block-free-u02-c07", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 87640, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 22.08, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1036800, "script_start_s": 0.0, "script_end_s": 21.6, "all_asr_last_word_s": 21.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8345545462457087, "source_start_sample": 0, "source_end_sample": 1044000, "first_word_output_sample": 0, "last_word_output_sample": 1036800, "samples": 1044000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.33, "added_silence_s": 0.0, "duration_s": 21.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u02-c07.mp3" }, "en-conversation-02-podcast-segment-free-u00-c00": { "id": "en-conversation-02-podcast-segment-free-u00-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74101, "plain": "Imagine a tiny radio studio where the hosts have just discovered this voice model.", "source_duration_s": 15.76, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 318720, "script_start_s": 0.24, "script_end_s": 6.64, "all_asr_last_word_s": 13.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.6428571428571429, "source_identity_reward": 0.26757298548791064, "source_start_sample": 7680, "source_end_sample": 325920, "first_word_output_sample": 3840, "last_word_output_sample": 311040, "samples": 318240, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 8.97, "added_silence_s": 0.0, "duration_s": 6.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-free-u00-c00.mp3" }, "de-conversation-02-podcast-segment-timed-u07-c00": { "id": "de-conversation-02-podcast-segment-timed-u07-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84213, "plain": "Damit kann ich etwas anfangen. Und die kleine rote Lampe bleibt bitte an.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 222720, "script_start_s": 0.08, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-timed-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.48847636719179943, "source_start_sample": 0, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-timed-u07-c00.mp3" }, "en-conversation-03-podcast-segment-timed-u02-c00": { "id": "en-conversation-03-podcast-segment-timed-u02-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75133, "plain": "Yes. The model sees speaker-labelled references and, when available, earlier dialogue.", "source_duration_s": 3.84, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 10, "coverage": 0.8333333333333334, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.16666666666666666, "source_identity_reward": 0.36534187241896016 }, "en-conversation-01-podcast-segment-timed-u04-c00": { "id": "en-conversation-01-podcast-segment-timed-u04-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 73165, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 10, "coverage": 0.8333333333333334, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 192000, "script_start_s": 0.16, "script_end_s": 4.0, "all_asr_last_word_s": 4.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.16666666666666666, "source_identity_reward": 0.26004262327246225, "source_start_sample": 3840, "source_end_sample": 195840, "first_word_output_sample": 3840, "last_word_output_sample": 188160, "samples": 195360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-timed-u04-c00.mp3" }, "en-conversation-05-podcast-segment-free-u14-c00": { "id": "en-conversation-05-podcast-segment-free-u14-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77325, "plain": "Exactly. The exchange should feel responsive, with the same two people returning after every change of mood.", "source_duration_s": 15.2, "source_frame_cap": false, "alignment": { "expected_tokens": 17, "matched_tokens": 17, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 295680, "script_start_s": 0.24, "script_end_s": 6.16, "all_asr_last_word_s": 14.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u14-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.8235294117647058, "source_identity_reward": 0.22806453466206486, "source_start_sample": 7680, "source_end_sample": 302880, "first_word_output_sample": 3840, "last_word_output_sample": 288000, "samples": 295200, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 8.89, "added_silence_s": 0.0, "duration_s": 6.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-free-u14-c00.mp3" }, "en-conversation-05-podcast-block-free-u00-c00": { "id": "en-conversation-05-podcast-block-free-u00-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 77601, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 21.04, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 975360, "script_start_s": 0.16, "script_end_s": 20.32, "all_asr_last_word_s": 20.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5497311827789649, "source_start_sample": 3840, "source_end_sample": 982560, "first_word_output_sample": 3840, "last_word_output_sample": 971520, "samples": 978720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.57, "added_silence_s": 0.0, "duration_s": 20.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u00-c00.mp3" }, "en-conversation-05-podcast-block-free-u00-c01": { "id": "en-conversation-05-podcast-block-free-u00-c01", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 77602, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 19.84, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 948480, "script_start_s": 0.16, "script_end_s": 19.76, "all_asr_last_word_s": 19.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.6170544932415828, "source_start_sample": 3840, "source_end_sample": 952320, "first_word_output_sample": 3840, "last_word_output_sample": 944640, "samples": 951840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 19.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u00-c01.mp3" }, "en-conversation-05-podcast-block-free-u00-c02": { "id": "en-conversation-05-podcast-block-free-u00-c02", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 77603, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 20.24, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 967680, "script_start_s": 0.0, "script_end_s": 20.16, "all_asr_last_word_s": 20.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.7182708912585715, "source_start_sample": 0, "source_end_sample": 971520, "first_word_output_sample": 0, "last_word_output_sample": 967680, "samples": 974880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 20.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u00-c02.mp3" }, "en-conversation-05-podcast-block-free-u00-c03": { "id": "en-conversation-05-podcast-block-free-u00-c03", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 77604, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 961920, "script_start_s": 0.0, "script_end_s": 20.04, "all_asr_last_word_s": 20.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.6188660651991037, "source_start_sample": 0, "source_end_sample": 969120, "first_word_output_sample": 0, "last_word_output_sample": 961920, "samples": 969120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 39.81, "added_silence_s": 0.0, "duration_s": 20.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u00-c03.mp3" }, "en-conversation-05-podcast-block-free-u00-c04": { "id": "en-conversation-05-podcast-block-free-u00-c04", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 77605, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 25.12, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 921600, "script_start_s": 0.0, "script_end_s": 19.2, "all_asr_last_word_s": 24.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3090909090909091, "source_identity_reward": 0.387387118259914, "source_start_sample": 0, "source_end_sample": 928800, "first_word_output_sample": 0, "last_word_output_sample": 921600, "samples": 928800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 5.77, "added_silence_s": 0.0, "duration_s": 19.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u00-c04.mp3" }, "en-conversation-05-podcast-block-free-u00-c05": { "id": "en-conversation-05-podcast-block-free-u00-c05", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 77606, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 21.28, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 52, "coverage": 0.9454545454545454, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 956160, "script_start_s": 0.08, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05454545454545454, "source_identity_reward": 0.5275064887462295, "source_start_sample": 0, "source_end_sample": 963360, "first_word_output_sample": 3840, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.21, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u00-c05.mp3" }, "en-conversation-05-podcast-block-free-u00-c06": { "id": "en-conversation-05-podcast-block-free-u00-c06", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 77607, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 22.64, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 971520, "script_start_s": 0.16, "script_end_s": 20.24, "all_asr_last_word_s": 20.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.6442308246082393, "source_start_sample": 3840, "source_end_sample": 978720, "first_word_output_sample": 3840, "last_word_output_sample": 967680, "samples": 974880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 2.25, "added_silence_s": 0.0, "duration_s": 20.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u00-c06.mp3" }, "en-conversation-05-podcast-block-free-u00-c07": { "id": "en-conversation-05-podcast-block-free-u00-c07", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 77608, "plain": "Picture a listener hearing an English exchange, then German, French, Spanish and Dutch. Same model family, different voices, and a different emotional rhythm in each language? That is what this five-language grid explores, with short and longer conversations. I would start with no references and ask it to invent two hosts. Risky, but fascinating.", "source_duration_s": 20.48, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 960000, "script_start_s": 0.16, "script_end_s": 20.0, "all_asr_last_word_s": 20.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.5653504166491543, "source_start_sample": 3840, "source_end_sample": 967200, "first_word_output_sample": 3840, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.33, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-free-u00-c07.mp3" }, "de-conversation-02-podcast-segment-free-u05-c00": { "id": "de-conversation-02-podcast-segment-free-u05-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84181, "plain": "Von einer begeisterten Beschreibung lasse ich mich noch lange nicht überzeugen. Ich will es hören.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 276480, "script_start_s": 0.0, "script_end_s": 5.76, "all_asr_last_word_s": 8.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2, "source_identity_reward": 0.5107990411115987, "source_start_sample": 0, "source_end_sample": 283680, "first_word_output_sample": 0, "last_word_output_sample": 276480, "samples": 283680, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.09, "added_silence_s": 0.0, "duration_s": 5.91, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-free-u05-c00.mp3" }, "en-conversation-04-podcast-segment-free-u03-c00": { "id": "en-conversation-04-podcast-segment-free-u03-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76149, "plain": "Thank you. I was about to ask it to fit my entire life into three seconds.", "source_duration_s": 9.28, "source_frame_cap": false, "alignment": { "expected_tokens": 16, "matched_tokens": 16, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 241920, "script_start_s": 0.16, "script_end_s": 5.04, "all_asr_last_word_s": 5.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.13613434524653967, "source_start_sample": 3840, "source_end_sample": 249120, "first_word_output_sample": 3840, "last_word_output_sample": 238080, "samples": 245280, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 4.09, "added_silence_s": 0.0, "duration_s": 5.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-free-u03-c00.mp3" }, "de-conversation-05-podcast-segment-free-u03-c00": { "id": "de-conversation-05-podcast-segment-free-u03-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87149, "plain": "Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 322560, "script_start_s": 0.16, "script_end_s": 6.72, "all_asr_last_word_s": 29.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.4615384615384615, "source_identity_reward": 0.20798497975532262, "source_start_sample": 3840, "source_end_sample": 329760, "first_word_output_sample": 3840, "last_word_output_sample": 318720, "samples": 325920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.13, "added_silence_s": 0.0, "duration_s": 6.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u03-c00.mp3" }, "en-conversation-03-podcast-block-timed-u02-c00": { "id": "en-conversation-03-podcast-block-timed-u02-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 75633, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 875520, "script_start_s": 0.16, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4601129624322676, "source_start_sample": 3840, "source_end_sample": 879360, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u02-c00.mp3" }, "en-conversation-03-podcast-block-timed-u02-c01": { "id": "en-conversation-03-podcast-block-timed-u02-c01", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 75634, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 875520, "script_start_s": 0.16, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.4486960845985569, "source_start_sample": 3840, "source_end_sample": 879360, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u02-c01.mp3" }, "en-conversation-03-podcast-block-timed-u02-c02": { "id": "en-conversation-03-podcast-block-timed-u02-c02", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 75635, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.48, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 53, "coverage": 0.9636363636363636, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03636363636363636, "source_identity_reward": 0.42199881996272864 }, "en-conversation-03-podcast-block-timed-u02-c03": { "id": "en-conversation-03-podcast-block-timed-u02-c03", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 75636, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 875520, "script_start_s": 0.0, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.4495885456299368, "source_start_sample": 0, "source_end_sample": 879360, "first_word_output_sample": 0, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u02-c03.mp3" }, "en-conversation-03-podcast-block-timed-u02-c04": { "id": "en-conversation-03-podcast-block-timed-u02-c04", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 75637, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 5.12, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 14, "coverage": 0.2545454545454545, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": false, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.7454545454545455, "source_identity_reward": 0.09154299980847644 }, "en-conversation-03-podcast-block-timed-u02-c05": { "id": "en-conversation-03-podcast-block-timed-u02-c05", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 75638, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 875520, "script_start_s": 0.16, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4770332817550139, "source_start_sample": 3840, "source_end_sample": 879360, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u02-c05.mp3" }, "en-conversation-03-podcast-block-timed-u02-c06": { "id": "en-conversation-03-podcast-block-timed-u02-c06", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 75639, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 875520, "script_start_s": 0.16, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5018693509470453, "source_start_sample": 3840, "source_end_sample": 879360, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u02-c06.mp3" }, "en-conversation-03-podcast-block-timed-u02-c07": { "id": "en-conversation-03-podcast-block-timed-u02-c07", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 75640, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 875520, "script_start_s": 0.16, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6629703038789412, "source_start_sample": 3840, "source_end_sample": 879360, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u02-c07.mp3" }, "de-conversation-04-podcast-block-timed-u00-c00": { "id": "de-conversation-04-podcast-block-timed-u00-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 86601, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 59, "coverage": 0.9365079365079365, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06349206349206349, "source_identity_reward": 0.7382352828659238 }, "de-conversation-04-podcast-block-timed-u00-c01": { "id": "de-conversation-04-podcast-block-timed-u00-c01", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 86602, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.68, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 57, "coverage": 0.9047619047619048, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09523809523809523, "source_identity_reward": 0.5942625546459879 }, "de-conversation-04-podcast-block-timed-u00-c02": { "id": "de-conversation-04-podcast-block-timed-u00-c02", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 86603, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 61, "coverage": 0.9682539682539683, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1040640, "script_start_s": 0.0, "script_end_s": 21.68, "all_asr_last_word_s": 21.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.031746031746031744, "source_identity_reward": 0.6481860692658551, "source_start_sample": 0, "source_end_sample": 1044480, "first_word_output_sample": 0, "last_word_output_sample": 1040640, "samples": 1047840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u00-c02.mp3" }, "de-conversation-04-podcast-block-timed-u00-c03": { "id": "de-conversation-04-podcast-block-timed-u00-c03", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 86604, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 58, "coverage": 0.9206349206349206, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1040640, "script_start_s": 0.0, "script_end_s": 21.68, "all_asr_last_word_s": 21.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07936507936507936, "source_identity_reward": 0.6975511347508224, "source_start_sample": 0, "source_end_sample": 1044480, "first_word_output_sample": 0, "last_word_output_sample": 1040640, "samples": 1047840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u00-c03.mp3" }, "de-conversation-04-podcast-block-timed-u00-c04": { "id": "de-conversation-04-podcast-block-timed-u00-c04", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 86605, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 59, "coverage": 0.9365079365079365, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1040640, "script_start_s": 0.0, "script_end_s": 21.68, "all_asr_last_word_s": 21.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06349206349206349, "source_identity_reward": 0.6332172368926536, "source_start_sample": 0, "source_end_sample": 1044480, "first_word_output_sample": 0, "last_word_output_sample": 1040640, "samples": 1047840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u00-c04.mp3" }, "de-conversation-04-podcast-block-timed-u00-c05": { "id": "de-conversation-04-podcast-block-timed-u00-c05", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 86606, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 59, "coverage": 0.9365079365079365, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06349206349206349, "source_identity_reward": 0.6547140969908886 }, "de-conversation-04-podcast-block-timed-u00-c06": { "id": "de-conversation-04-podcast-block-timed-u00-c06", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 86607, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.84, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 57, "coverage": 0.9047619047619048, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.09523809523809523, "source_identity_reward": 0.7229257576252657 }, "de-conversation-04-podcast-block-timed-u00-c07": { "id": "de-conversation-04-podcast-block-timed-u00-c07", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 86608, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben. Du machst Witze. Trifft das Modell dann wirklich jede einzelne Millisekunde? Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden. Danke. Ich wollte gerade mein ganzes Leben auf drei Sekunden zusammenfassen lassen. Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 61, "coverage": 0.9682539682539683, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1040640, "script_start_s": 0.0, "script_end_s": 21.68, "all_asr_last_word_s": 21.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.031746031746031744, "source_identity_reward": 0.6467924238211564, "source_start_sample": 0, "source_end_sample": 1044480, "first_word_output_sample": 0, "last_word_output_sample": 1040640, "samples": 1047840, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 21.83, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u00-c07.mp3" }, "de-conversation-04-podcast-segment-free-u04-c00": { "id": "de-conversation-04-podcast-segment-free-u04-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86165, "plain": "Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 11.2, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 307200, "script_start_s": 0.0, "script_end_s": 6.4, "all_asr_last_word_s": 7.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.15384615384615385, "source_identity_reward": 0.5846798442816443, "source_start_sample": 0, "source_end_sample": 314400, "first_word_output_sample": 0, "last_word_output_sample": 307200, "samples": 314400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.65, "added_silence_s": 0.0, "duration_s": 6.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-free-u04-c00.mp3" }, "en-conversation-05-podcast-segment-timed-u06-c00": { "id": "en-conversation-05-podcast-segment-timed-u06-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77197, "plain": "Exactly. We publish the full scripts and seeds so the exciting failures remain visible too.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 13, "coverage": 0.8666666666666667, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.13333333333333333, "source_identity_reward": 0.3978047704476744 }, "en-conversation-01-podcast-block-timed-u01-c00": { "id": "en-conversation-01-podcast-block-timed-u01-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 73617, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.76, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 844800, "script_start_s": 0.0, "script_end_s": 17.6, "all_asr_last_word_s": 17.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.531585547443013, "source_start_sample": 0, "source_end_sample": 852000, "first_word_output_sample": 0, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u01-c00.mp3" }, "en-conversation-01-podcast-block-timed-u01-c01": { "id": "en-conversation-01-podcast-block-timed-u01-c01", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 73618, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.68, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 844800, "script_start_s": 0.16, "script_end_s": 17.6, "all_asr_last_word_s": 17.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.45171417638771805, "source_start_sample": 3840, "source_end_sample": 848640, "first_word_output_sample": 3840, "last_word_output_sample": 840960, "samples": 848160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u01-c01.mp3" }, "en-conversation-01-podcast-block-timed-u01-c02": { "id": "en-conversation-01-podcast-block-timed-u01-c02", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 73619, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.68, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 840960, "script_start_s": 0.0, "script_end_s": 17.52, "all_asr_last_word_s": 17.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.3613600626066998, "source_start_sample": 0, "source_end_sample": 848160, "first_word_output_sample": 0, "last_word_output_sample": 840960, "samples": 848160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u01-c02.mp3" }, "en-conversation-01-podcast-block-timed-u01-c03": { "id": "en-conversation-01-podcast-block-timed-u01-c03", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 73620, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.68, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 844800, "script_start_s": 0.0, "script_end_s": 17.6, "all_asr_last_word_s": 17.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.35586170857553706, "source_start_sample": 0, "source_end_sample": 848640, "first_word_output_sample": 0, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u01-c03.mp3" }, "en-conversation-01-podcast-block-timed-u01-c04": { "id": "en-conversation-01-podcast-block-timed-u01-c04", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 73621, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.68, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 844800, "script_start_s": 0.0, "script_end_s": 17.6, "all_asr_last_word_s": 17.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.3730122825480249, "source_start_sample": 0, "source_end_sample": 848640, "first_word_output_sample": 0, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u01-c04.mp3" }, "en-conversation-01-podcast-block-timed-u01-c05": { "id": "en-conversation-01-podcast-block-timed-u01-c05", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 73622, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.68, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 844800, "script_start_s": 0.0, "script_end_s": 17.6, "all_asr_last_word_s": 17.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.44830527372277296, "source_start_sample": 0, "source_end_sample": 848640, "first_word_output_sample": 0, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u01-c05.mp3" }, "en-conversation-01-podcast-block-timed-u01-c06": { "id": "en-conversation-01-podcast-block-timed-u01-c06", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 73623, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.68, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 50, "coverage": 0.9433962264150944, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 840960, "script_start_s": 0.16, "script_end_s": 17.52, "all_asr_last_word_s": 17.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05660377358490566, "source_identity_reward": 0.5023673003892499, "source_start_sample": 3840, "source_end_sample": 848160, "first_word_output_sample": 3840, "last_word_output_sample": 837120, "samples": 844320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u01-c06.mp3" }, "en-conversation-01-podcast-block-timed-u01-c07": { "id": "en-conversation-01-podcast-block-timed-u01-c07", "case": "en-conversation-01", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 73624, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances. That distinction makes me trust the demonstration a little more. Annoyingly sensible of you. The new training mixes real conversations with acting examples to retain both formats. All right, show me the awkward pause before someone finally admits they are impressed.", "source_duration_s": 17.68, "source_frame_cap": false, "alignment": { "expected_tokens": 53, "matched_tokens": 49, "coverage": 0.9245283018867925, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 844800, "script_start_s": 0.0, "script_end_s": 17.6, "all_asr_last_word_s": 17.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07547169811320754, "source_identity_reward": 0.4265848686035359, "source_start_sample": 0, "source_end_sample": 848640, "first_word_output_sample": 0, "last_word_output_sample": 844800, "samples": 852000, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.75, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-block-timed-u01-c07.mp3" }, "de-conversation-05-podcast-segment-free-u09-c00": { "id": "de-conversation-05-podcast-segment-free-u09-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 349440, "script_start_s": 0.16, "script_end_s": 7.28, "all_asr_last_word_s": 28.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.1428571428571428, "source_identity_reward": 0.19686396033734802, "source_start_sample": 3840, "source_end_sample": 356640, "first_word_output_sample": 3840, "last_word_output_sample": 345600, "samples": 352800, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 22.57, "added_silence_s": 0.0, "duration_s": 7.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u09-c00.mp3" }, "de-conversation-04-podcast-segment-timed-u00-c00": { "id": "de-conversation-04-podcast-segment-timed-u00-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86101, "plain": "Im Skript kann ich neben einen einzelnen Satz eine gewünschte Dauer schreiben.", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 207360, "script_start_s": 0.0, "script_end_s": 4.32, "all_asr_last_word_s": 4.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7068620056718933, "source_start_sample": 0, "source_end_sample": 211200, "first_word_output_sample": 0, "last_word_output_sample": 207360, "samples": 214560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-timed-u00-c00.mp3" }, "de-conversation-03-podcast-segment-free-u02-c00": { "id": "de-conversation-03-podcast-segment-free-u02-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85133, "plain": "Ja. Das Modell erhält markierte Sprecherreferenzen und manchmal auch den bisherigen Gesprächsverlauf.", "source_duration_s": 15.04, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 303360, "script_start_s": 0.16, "script_end_s": 6.32, "all_asr_last_word_s": 14.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.4166666666666667, "source_identity_reward": 0.22560516299195207, "source_start_sample": 3840, "source_end_sample": 310560, "first_word_output_sample": 3840, "last_word_output_sample": 299520, "samples": 306720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 8.57, "added_silence_s": 0.0, "duration_s": 6.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-free-u02-c00.mp3" }, "en-conversation-02-podcast-segment-timed-u03-c00": { "id": "en-conversation-02-podcast-segment-timed-u03-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74149, "plain": "And does the second host interrupt with healthy skepticism, or just politely agree?", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 203520, "script_start_s": 0.0, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.196277783210352, "source_start_sample": 0, "source_end_sample": 210720, "first_word_output_sample": 0, "last_word_output_sample": 203520, "samples": 210720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-timed-u03-c00.mp3" }, "de-conversation-02-podcast-block-timed-u02-c00": { "id": "de-conversation-02-podcast-block-timed-u02-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 84633, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 491520, "script_start_s": 0.0, "script_end_s": 10.24, "all_asr_last_word_s": 10.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7677140046640627, "source_start_sample": 0, "source_end_sample": 495360, "first_word_output_sample": 0, "last_word_output_sample": 491520, "samples": 498720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 10.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-timed-u02-c00.mp3" }, "de-conversation-02-podcast-block-timed-u02-c01": { "id": "de-conversation-02-podcast-block-timed-u02-c01", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 84634, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 28, "coverage": 0.9655172413793104, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 487680, "script_start_s": 0.0, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06896551724137931, "source_identity_reward": 0.7009521821026767, "source_start_sample": 0, "source_end_sample": 494880, "first_word_output_sample": 0, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-timed-u02-c01.mp3" }, "de-conversation-02-podcast-block-timed-u02-c02": { "id": "de-conversation-02-podcast-block-timed-u02-c02", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 84635, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 491520, "script_start_s": 0.0, "script_end_s": 10.24, "all_asr_last_word_s": 10.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7733266678852647, "source_start_sample": 0, "source_end_sample": 495360, "first_word_output_sample": 0, "last_word_output_sample": 491520, "samples": 498720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 10.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-timed-u02-c02.mp3" }, "de-conversation-02-podcast-block-timed-u02-c03": { "id": "de-conversation-02-podcast-block-timed-u02-c03", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 84636, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 28, "coverage": 0.9655172413793104, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 487680, "script_start_s": 0.0, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06896551724137931, "source_identity_reward": 0.6690220183685488, "source_start_sample": 0, "source_end_sample": 494880, "first_word_output_sample": 0, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-timed-u02-c03.mp3" }, "de-conversation-02-podcast-block-timed-u02-c04": { "id": "de-conversation-02-podcast-block-timed-u02-c04", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 84637, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 487680, "script_start_s": 0.0, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6775570525994639, "source_start_sample": 0, "source_end_sample": 494880, "first_word_output_sample": 0, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-timed-u02-c04.mp3" }, "de-conversation-02-podcast-block-timed-u02-c05": { "id": "de-conversation-02-podcast-block-timed-u02-c05", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 84638, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 487680, "script_start_s": 0.0, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7467102580020727, "source_start_sample": 0, "source_end_sample": 494880, "first_word_output_sample": 0, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-timed-u02-c05.mp3" }, "de-conversation-02-podcast-block-timed-u02-c06": { "id": "de-conversation-02-podcast-block-timed-u02-c06", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 84639, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 29, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 487680, "script_start_s": 0.0, "script_end_s": 10.16, "all_asr_last_word_s": 10.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7682971882713208, "source_start_sample": 0, "source_end_sample": 494880, "first_word_output_sample": 0, "last_word_output_sample": 487680, "samples": 494880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 10.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-block-timed-u02-c06.mp3" }, "de-conversation-02-podcast-block-timed-u02-c07": { "id": "de-conversation-02-podcast-block-timed-u02-c07", "case": "de-conversation-02", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 84640, "plain": "Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 29, "matched_tokens": 27, "coverage": 0.9310344827586207, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06896551724137931, "source_identity_reward": 0.6990225106693841 }, "de-conversation-01-podcast-segment-timed-u05-c00": { "id": "de-conversation-01-podcast-segment-timed-u05-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83181, "plain": "Diese Einschränkung macht die Vorführung irgendwie glaubwürdiger. Ärgerlich vernünftig von dir.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 11, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 192000, "script_start_s": 0.0, "script_end_s": 4.0, "all_asr_last_word_s": 4.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4779126867926934, "source_start_sample": 0, "source_end_sample": 195840, "first_word_output_sample": 0, "last_word_output_sample": 192000, "samples": 199200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-timed-u05-c00.mp3" }, "en-conversation-03-podcast-block-free-u00-c00": { "id": "en-conversation-03-podcast-block-free-u00-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 75601, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 19.2, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 51, "coverage": 0.9807692307692307, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 913920, "script_start_s": 0.0, "script_end_s": 19.04, "all_asr_last_word_s": 19.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.019230769230769232, "source_identity_reward": 0.5099945615619067, "source_start_sample": 0, "source_end_sample": 921120, "first_word_output_sample": 0, "last_word_output_sample": 913920, "samples": 921120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u00-c00.mp3" }, "en-conversation-03-podcast-block-free-u00-c01": { "id": "en-conversation-03-podcast-block-free-u00-c01", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 75602, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 19.04, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 51, "coverage": 0.9807692307692307, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 902400, "script_start_s": 0.16, "script_end_s": 18.8, "all_asr_last_word_s": 18.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.019230769230769232, "source_identity_reward": 0.515503580340004, "source_start_sample": 3840, "source_end_sample": 909600, "first_word_output_sample": 3840, "last_word_output_sample": 898560, "samples": 905760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 18.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u00-c01.mp3" }, "en-conversation-03-podcast-block-free-u00-c02": { "id": "en-conversation-03-podcast-block-free-u00-c02", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 75603, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 52, "matched_tokens": 51, "coverage": 0.9807692307692307, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 929280, "script_start_s": 0.16, "script_end_s": 19.36, "all_asr_last_word_s": 59.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.2307692307692308, "source_identity_reward": 0.26148914519823235, "source_start_sample": 3840, "source_end_sample": 936480, "first_word_output_sample": 3840, "last_word_output_sample": 925440, "samples": 932640, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 40.49, "added_silence_s": 0.0, "duration_s": 19.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u00-c02.mp3" }, "en-conversation-03-podcast-block-free-u00-c03": { "id": "en-conversation-03-podcast-block-free-u00-c03", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 75604, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 925440, "script_start_s": 0.0, "script_end_s": 19.28, "all_asr_last_word_s": 59.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.3653846153846154, "source_identity_reward": 0.2422114790174885, "source_start_sample": 0, "source_end_sample": 932640, "first_word_output_sample": 0, "last_word_output_sample": 925440, "samples": 932640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 40.57, "added_silence_s": 0.0, "duration_s": 19.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u00-c03.mp3" }, "en-conversation-03-podcast-block-free-u00-c04": { "id": "en-conversation-03-podcast-block-free-u00-c04", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 75605, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 18.48, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 51, "coverage": 0.9807692307692307, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 879360, "script_start_s": 0.0, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.019230769230769232, "source_identity_reward": 0.5205010664262022, "source_start_sample": 0, "source_end_sample": 886560, "first_word_output_sample": 0, "last_word_output_sample": 879360, "samples": 886560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 18.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u00-c04.mp3" }, "en-conversation-03-podcast-block-free-u00-c05": { "id": "en-conversation-03-podcast-block-free-u00-c05", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 75606, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 26.48, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 929280, "script_start_s": 0.08, "script_end_s": 19.36, "all_asr_last_word_s": 19.36 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.5439063409589739, "source_start_sample": 0, "source_end_sample": 936480, "first_word_output_sample": 3840, "last_word_output_sample": 929280, "samples": 936480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 6.97, "added_silence_s": 0.0, "duration_s": 19.51, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u00-c05.mp3" }, "en-conversation-03-podcast-block-free-u00-c06": { "id": "en-conversation-03-podcast-block-free-u00-c06", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 75607, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 19.44, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 51, "coverage": 0.9807692307692307, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 925440, "script_start_s": 0.0, "script_end_s": 19.28, "all_asr_last_word_s": 19.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.019230769230769232, "source_identity_reward": 0.5266339899325216, "source_start_sample": 0, "source_end_sample": 932640, "first_word_output_sample": 0, "last_word_output_sample": 925440, "samples": 932640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u00-c06.mp3" }, "en-conversation-03-podcast-block-free-u00-c07": { "id": "en-conversation-03-podcast-block-free-u00-c07", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 75608, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 19.44, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 51, "coverage": 0.9807692307692307, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 925440, "script_start_s": 0.0, "script_end_s": 19.28, "all_asr_last_word_s": 19.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.019230769230769232, "source_identity_reward": 0.5124244497913377, "source_start_sample": 0, "source_end_sample": 932640, "first_word_output_sample": 0, "last_word_output_sample": 925440, "samples": 932640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 19.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u00-c07.mp3" }, "en-conversation-02-podcast-segment-timed-u01-c00": { "id": "en-conversation-02-podcast-segment-timed-u01-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74117, "plain": "Please tell me the tiny studio has a very dramatic red recording light.", "source_duration_s": 4.4, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 203520, "script_start_s": 0.0, "script_end_s": 4.24, "all_asr_last_word_s": 4.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3944159026885496, "source_start_sample": 0, "source_end_sample": 210720, "first_word_output_sample": 0, "last_word_output_sample": 203520, "samples": 210720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-timed-u01-c00.mp3" }, "en-conversation-01-podcast-segment-free-u04-c00": { "id": "en-conversation-01-podcast-segment-free-u04-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73165, "plain": "Exactly. We are testing forty emotion labels, not claiming forty perfect performances.", "source_duration_s": 7.12, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 10, "coverage": 0.8333333333333334, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 272640, "script_start_s": 0.16, "script_end_s": 5.68, "all_asr_last_word_s": 5.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-free-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.16666666666666666, "source_identity_reward": 0.49158635106786297, "source_start_sample": 3840, "source_end_sample": 279840, "first_word_output_sample": 3840, "last_word_output_sample": 268800, "samples": 276000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.29, "added_silence_s": 0.0, "duration_s": 5.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-free-u04-c00.mp3" }, "en-conversation-03-podcast-segment-free-u05-c00": { "id": "en-conversation-03-podcast-segment-free-u05-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75181, "plain": "Good. I want the delighted host to remain the same person when she becomes doubtful.", "source_duration_s": 6.96, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 326400, "script_start_s": 0.16, "script_end_s": 6.8, "all_asr_last_word_s": 6.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.10657877126320277, "source_start_sample": 3840, "source_end_sample": 333600, "first_word_output_sample": 3840, "last_word_output_sample": 322560, "samples": 329760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 6.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-free-u05-c00.mp3" }, "en-conversation-01-podcast-segment-free-u08-c00": { "id": "en-conversation-01-podcast-segment-free-u08-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 264960, "script_start_s": 0.0, "script_end_s": 5.52, "all_asr_last_word_s": 29.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 2.7142857142857144, "source_identity_reward": 0.07791132059831532, "source_start_sample": 0, "source_end_sample": 272160, "first_word_output_sample": 0, "last_word_output_sample": 264960, "samples": 272160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 24.33, "added_silence_s": 0.0, "duration_s": 5.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-free-u08-c00.mp3" }, "en-conversation-01-podcast-segment-free-u11-c00": { "id": "en-conversation-01-podcast-segment-free-u11-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 10.8, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 253440, "script_start_s": 0.16, "script_end_s": 5.28, "all_asr_last_word_s": 5.28 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-free-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.15941306022713686, "source_start_sample": 3840, "source_end_sample": 260640, "first_word_output_sample": 3840, "last_word_output_sample": 249600, "samples": 256800, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 5.37, "added_silence_s": 0.0, "duration_s": 5.35, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-free-u11-c00.mp3" }, "en-conversation-04-podcast-segment-free-u02-c00": { "id": "en-conversation-04-podcast-segment-free-u02-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76133, "plain": "No. The codec advances in eighty-millisecond frames, and requested timing can still miss.", "source_duration_s": 8.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 13, "coverage": 0.9285714285714286, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 245760, "script_start_s": 0.0, "script_end_s": 5.12, "all_asr_last_word_s": 5.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07142857142857142, "source_identity_reward": 0.3640099980151058, "source_start_sample": 0, "source_end_sample": 252960, "first_word_output_sample": 0, "last_word_output_sample": 245760, "samples": 252960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 3.45, "added_silence_s": 0.0, "duration_s": 5.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-free-u02-c00.mp3" }, "de-conversation-03-podcast-block-free-u02-c00": { "id": "de-conversation-03-podcast-block-free-u02-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 85633, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.32, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 967680, "script_start_s": 0.0, "script_end_s": 20.16, "all_asr_last_word_s": 20.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.017857142857142856, "source_identity_reward": 0.8300450219137288, "source_start_sample": 0, "source_end_sample": 974880, "first_word_output_sample": 0, "last_word_output_sample": 967680, "samples": 974880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u02-c00.mp3" }, "de-conversation-03-podcast-block-free-u02-c01": { "id": "de-conversation-03-podcast-block-free-u02-c01", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 85634, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 23.44, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1025280, "script_start_s": 0.0, "script_end_s": 21.36, "all_asr_last_word_s": 23.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05357142857142857, "source_identity_reward": 0.7744970368117179, "source_start_sample": 0, "source_end_sample": 1032480, "first_word_output_sample": 0, "last_word_output_sample": 1025280, "samples": 1032480, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.93, "added_silence_s": 0.0, "duration_s": 21.51, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u02-c01.mp3" }, "de-conversation-03-podcast-block-free-u02-c02": { "id": "de-conversation-03-podcast-block-free-u02-c02", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 85635, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.72, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 986880, "script_start_s": 0.0, "script_end_s": 20.56, "all_asr_last_word_s": 20.56 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7822221260023051, "source_start_sample": 0, "source_end_sample": 994080, "first_word_output_sample": 0, "last_word_output_sample": 986880, "samples": 994080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u02-c02.mp3" }, "de-conversation-03-podcast-block-free-u02-c03": { "id": "de-conversation-03-podcast-block-free-u02-c03", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 85636, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 55, "coverage": 0.9821428571428571, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1021440, "script_start_s": 0.0, "script_end_s": 21.28, "all_asr_last_word_s": 59.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.7142857142857143, "source_identity_reward": 0.4676254246046107, "source_start_sample": 0, "source_end_sample": 1028640, "first_word_output_sample": 0, "last_word_output_sample": 1021440, "samples": 1028640, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.57, "added_silence_s": 0.0, "duration_s": 21.43, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u02-c03.mp3" }, "de-conversation-03-podcast-block-free-u02-c04": { "id": "de-conversation-03-podcast-block-free-u02-c04", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 85637, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.32, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 967680, "script_start_s": 0.0, "script_end_s": 20.16, "all_asr_last_word_s": 20.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7877532379758756, "source_start_sample": 0, "source_end_sample": 974880, "first_word_output_sample": 0, "last_word_output_sample": 967680, "samples": 974880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u02-c04.mp3" }, "de-conversation-03-podcast-block-free-u02-c05": { "id": "de-conversation-03-podcast-block-free-u02-c05", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 85638, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 21.28, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 53, "coverage": 0.9464285714285714, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1013760, "script_start_s": 0.0, "script_end_s": 21.12, "all_asr_last_word_s": 21.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.05357142857142857, "source_identity_reward": 0.8153731431982806, "source_start_sample": 0, "source_end_sample": 1020960, "first_word_output_sample": 0, "last_word_output_sample": 1013760, "samples": 1020960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u02-c05.mp3" }, "de-conversation-03-podcast-block-free-u02-c06": { "id": "de-conversation-03-podcast-block-free-u02-c06", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 85639, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1013760, "script_start_s": 0.0, "script_end_s": 21.12, "all_asr_last_word_s": 57.480000000000004 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.2857142857142858, "source_identity_reward": 0.36250206367043414, "source_start_sample": 0, "source_end_sample": 1020960, "first_word_output_sample": 0, "last_word_output_sample": 1013760, "samples": 1020960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 38.73, "added_silence_s": 0.0, "duration_s": 21.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u02-c06.mp3" }, "de-conversation-03-podcast-block-free-u02-c07": { "id": "de-conversation-03-podcast-block-free-u02-c07", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 85640, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz. Dann hören wir genau hin, auch an den Stellen, an denen die schöne Illusion kurz zerbricht. Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 20.56, "source_frame_cap": false, "alignment": { "expected_tokens": 56, "matched_tokens": 56, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 979200, "script_start_s": 0.0, "script_end_s": 20.4, "all_asr_last_word_s": 20.400000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7526657404531292, "source_start_sample": 0, "source_end_sample": 986400, "first_word_output_sample": 0, "last_word_output_sample": 979200, "samples": 986400, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.55, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u02-c07.mp3" }, "de-conversation-03-podcast-segment-timed-u11-c00": { "id": "de-conversation-03-podcast-segment-timed-u11-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 222720, "script_start_s": 0.24, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6014128221188834, "source_start_sample": 7680, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 215040, "samples": 222240, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-timed-u11-c00.mp3" }, "de-conversation-03-podcast-segment-free-u00-c00": { "id": "de-conversation-03-podcast-segment-free-u00-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85101, "plain": "Mich interessiert besonders, ob kurze Aufnahmen die Identität beider Stimmen festlegen können.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 307200, "script_start_s": 0.16, "script_end_s": 6.4, "all_asr_last_word_s": 6.4 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5080545516423487, "source_start_sample": 3840, "source_end_sample": 314400, "first_word_output_sample": 3840, "last_word_output_sample": 303360, "samples": 310560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 23.45, "added_silence_s": 0.0, "duration_s": 6.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-free-u00-c00.mp3" }, "de-conversation-05-podcast-block-free-u01-c00": { "id": "de-conversation-05-podcast-block-free-u01-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 87617, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 23.6, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1048320, "script_start_s": 0.16, "script_end_s": 21.84, "all_asr_last_word_s": 21.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7947832079375519, "source_start_sample": 3840, "source_end_sample": 1055520, "first_word_output_sample": 3840, "last_word_output_sample": 1044480, "samples": 1051680, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.61, "added_silence_s": 0.0, "duration_s": 21.91, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u01-c00.mp3" }, "de-conversation-05-podcast-block-free-u01-c01": { "id": "de-conversation-05-podcast-block-free-u01-c01", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 87618, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 21.84, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1017600, "script_start_s": 0.16, "script_end_s": 21.2, "all_asr_last_word_s": 21.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7819515011259341, "source_start_sample": 3840, "source_end_sample": 1024800, "first_word_output_sample": 3840, "last_word_output_sample": 1013760, "samples": 1020960, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.49, "added_silence_s": 0.0, "duration_s": 21.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u01-c01.mp3" }, "de-conversation-05-podcast-block-free-u01-c02": { "id": "de-conversation-05-podcast-block-free-u01-c02", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 87619, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 22.8, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1032960, "script_start_s": 0.16, "script_end_s": 21.52, "all_asr_last_word_s": 21.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7438239302287137, "source_start_sample": 3840, "source_end_sample": 1040160, "first_word_output_sample": 3840, "last_word_output_sample": 1029120, "samples": 1036320, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.13, "added_silence_s": 0.0, "duration_s": 21.59, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u01-c02.mp3" }, "de-conversation-05-podcast-block-free-u01-c03": { "id": "de-conversation-05-podcast-block-free-u01-c03", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 87620, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1036800, "script_start_s": 0.16, "script_end_s": 21.6, "all_asr_last_word_s": 21.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7447785228325134, "source_start_sample": 3840, "source_end_sample": 1044000, "first_word_output_sample": 3840, "last_word_output_sample": 1032960, "samples": 1040160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u01-c03.mp3" }, "de-conversation-05-podcast-block-free-u01-c04": { "id": "de-conversation-05-podcast-block-free-u01-c04", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 87621, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 28.64, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 53, "coverage": 0.9814814814814815, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1013760, "script_start_s": 0.08, "script_end_s": 21.12, "all_asr_last_word_s": 25.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.14814814814814814, "source_identity_reward": 0.7009506126442727, "source_start_sample": 0, "source_end_sample": 1020960, "first_word_output_sample": 3840, "last_word_output_sample": 1013760, "samples": 1020960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 7.37, "added_silence_s": 0.0, "duration_s": 21.27, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u01-c04.mp3" }, "de-conversation-05-podcast-block-free-u01-c05": { "id": "de-conversation-05-podcast-block-free-u01-c05", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 87622, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 22.32, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1063680, "script_start_s": 0.16, "script_end_s": 22.16, "all_asr_last_word_s": 22.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6965745108140927, "source_start_sample": 3840, "source_end_sample": 1070880, "first_word_output_sample": 3840, "last_word_output_sample": 1059840, "samples": 1067040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u01-c05.mp3" }, "de-conversation-05-podcast-block-free-u01-c06": { "id": "de-conversation-05-podcast-block-free-u01-c06", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 87623, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 29.6, "source_frame_cap": false, "alignment": { "expected_tokens": 54, "matched_tokens": 52, "coverage": 0.9629629629629629, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1052160, "script_start_s": 0.08, "script_end_s": 21.92, "all_asr_last_word_s": 29.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2777777777777778, "source_identity_reward": 0.6703809101259551, "source_start_sample": 0, "source_end_sample": 1059360, "first_word_output_sample": 3840, "last_word_output_sample": 1052160, "samples": 1059360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 7.53, "added_silence_s": 0.0, "duration_s": 22.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u01-c06.mp3" }, "de-conversation-05-podcast-block-free-u01-c07": { "id": "de-conversation-05-podcast-block-free-u01-c07", "case": "de-conversation-05", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 87624, "plain": "Danach drei, klar als Versuch gekennzeichnet, weil das Training auf zwei Sprecher ausgerichtet war. Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen. Die vollständigen Skripte und Seeds bleiben sichtbar, damit auch interessante Fehler nachvollziehbar sind. Gut. Ich bin noch skeptisch, aber die nächste Antwort möchte ich jetzt wirklich hören.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 54, "matched_tokens": 54, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1105920, "script_start_s": 0.0, "script_end_s": 23.04, "all_asr_last_word_s": 53.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.5555555555555556, "source_identity_reward": 0.4310186569807053, "source_start_sample": 0, "source_end_sample": 1113120, "first_word_output_sample": 0, "last_word_output_sample": 1105920, "samples": 1113120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 36.81, "added_silence_s": 0.0, "duration_s": 23.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-block-free-u01-c07.mp3" }, "en-conversation-05-podcast-segment-timed-u13-c00": { "id": "en-conversation-05-podcast-segment-timed-u13-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77309, "plain": "And the second host is allowed to say, hold on, explain that again, I have not followed you.", "source_duration_s": 5.92, "source_frame_cap": false, "alignment": { "expected_tokens": 18, "matched_tokens": 18, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 276480, "script_start_s": 0.16, "script_end_s": 5.76, "all_asr_last_word_s": 5.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u13-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2104057750198488, "source_start_sample": 3840, "source_end_sample": 283680, "first_word_output_sample": 3840, "last_word_output_sample": 272640, "samples": 279840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 5.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-timed-u13-c00.mp3" }, "de-conversation-02-podcast-segment-free-u03-c00": { "id": "de-conversation-02-podcast-segment-free-u03-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84149, "plain": "Darf die zweite Stimme auch skeptisch nachfragen, statt immer nur freundlich zuzustimmen?", "source_duration_s": 10.48, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 288000, "script_start_s": 0.08, "script_end_s": 6.0, "all_asr_last_word_s": 6.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5036875405090735, "source_start_sample": 0, "source_end_sample": 295200, "first_word_output_sample": 3840, "last_word_output_sample": 288000, "samples": 295200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.33, "added_silence_s": 0.0, "duration_s": 6.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-free-u03-c00.mp3" }, "de-conversation-05-podcast-segment-free-u13-c00": { "id": "de-conversation-05-podcast-segment-free-u13-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87309, "plain": "Und die zweite Person darf sagen: Moment, erklär mir das bitte noch einmal, ich habe den Gedanken noch nicht verstanden.", "source_duration_s": 12.32, "source_frame_cap": false, "alignment": { "expected_tokens": 20, "matched_tokens": 20, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 314880, "script_start_s": 0.16, "script_end_s": 6.56, "all_asr_last_word_s": 11.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u13-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.35, "source_identity_reward": 0.34674426832569744, "source_start_sample": 3840, "source_end_sample": 322080, "first_word_output_sample": 3840, "last_word_output_sample": 311040, "samples": 318240, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 5.61, "added_silence_s": 0.0, "duration_s": 6.63, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u13-c00.mp3" }, "en-conversation-03-podcast-block-timed-u00-c00": { "id": "en-conversation-03-podcast-block-timed-u00-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 75601, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 17.04, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 51, "coverage": 0.9807692307692307, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 814080, "script_start_s": 0.0, "script_end_s": 16.96, "all_asr_last_word_s": 16.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.019230769230769232, "source_identity_reward": 0.46613742984644446, "source_start_sample": 0, "source_end_sample": 817920, "first_word_output_sample": 0, "last_word_output_sample": 814080, "samples": 821280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u00-c00.mp3" }, "en-conversation-03-podcast-block-timed-u00-c01": { "id": "en-conversation-03-podcast-block-timed-u00-c01", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 75602, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 17.04, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 810240, "script_start_s": 0.0, "script_end_s": 16.88, "all_asr_last_word_s": 16.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u00-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.4527293251713127, "source_start_sample": 0, "source_end_sample": 817440, "first_word_output_sample": 0, "last_word_output_sample": 810240, "samples": 817440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u00-c01.mp3" }, "en-conversation-03-podcast-block-timed-u00-c02": { "id": "en-conversation-03-podcast-block-timed-u00-c02", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 75603, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 17.04, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 810240, "script_start_s": 0.0, "script_end_s": 16.88, "all_asr_last_word_s": 16.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u00-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.5536170431466567, "source_start_sample": 0, "source_end_sample": 817440, "first_word_output_sample": 0, "last_word_output_sample": 810240, "samples": 817440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u00-c02.mp3" }, "en-conversation-03-podcast-block-timed-u00-c03": { "id": "en-conversation-03-podcast-block-timed-u00-c03", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 75604, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 17.04, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 49, "coverage": 0.9423076923076923, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 810240, "script_start_s": 0.0, "script_end_s": 16.88, "all_asr_last_word_s": 16.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u00-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.057692307692307696, "source_identity_reward": 0.4270696099709427, "source_start_sample": 0, "source_end_sample": 817440, "first_word_output_sample": 0, "last_word_output_sample": 810240, "samples": 817440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u00-c03.mp3" }, "en-conversation-03-podcast-block-timed-u00-c04": { "id": "en-conversation-03-podcast-block-timed-u00-c04", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 75605, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 17.04, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 810240, "script_start_s": 0.0, "script_end_s": 16.88, "all_asr_last_word_s": 16.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u00-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.5810070133720235, "source_start_sample": 0, "source_end_sample": 817440, "first_word_output_sample": 0, "last_word_output_sample": 810240, "samples": 817440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 17.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u00-c04.mp3" }, "en-conversation-03-podcast-block-timed-u00-c05": { "id": "en-conversation-03-podcast-block-timed-u00-c05", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 75606, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 17.04, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 814080, "script_start_s": 0.0, "script_end_s": 16.96, "all_asr_last_word_s": 16.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u00-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.5269079657565267, "source_start_sample": 0, "source_end_sample": 817920, "first_word_output_sample": 0, "last_word_output_sample": 814080, "samples": 821280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u00-c05.mp3" }, "en-conversation-03-podcast-block-timed-u00-c06": { "id": "en-conversation-03-podcast-block-timed-u00-c06", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 75607, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 17.04, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 814080, "script_start_s": 0.0, "script_end_s": 16.96, "all_asr_last_word_s": 16.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u00-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.5104640940169743, "source_start_sample": 0, "source_end_sample": 817920, "first_word_output_sample": 0, "last_word_output_sample": 814080, "samples": 821280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u00-c06.mp3" }, "en-conversation-03-podcast-block-timed-u00-c07": { "id": "en-conversation-03-podcast-block-timed-u00-c07", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 75608, "plain": "Here is the part that interests me: using a short recording to establish each voice. Then keeping those voices distinct while the conversation wanders somewhere unexpected? Yes. The model sees speaker-labelled references and, when available, earlier dialogue. I would love that to work, but long conversations are where voices can drift.", "source_duration_s": 16.96, "source_frame_cap": false, "alignment": { "expected_tokens": 52, "matched_tokens": 50, "coverage": 0.9615384615384616, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 810240, "script_start_s": 0.0, "script_end_s": 16.88, "all_asr_last_word_s": 16.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-timed-u00-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.038461538461538464, "source_identity_reward": 0.5129610710438353, "source_start_sample": 0, "source_end_sample": 814080, "first_word_output_sample": 0, "last_word_output_sample": 810240, "samples": 817440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 17.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-timed-u00-c07.mp3" }, "en-conversation-04-podcast-segment-timed-u06-c00": { "id": "en-conversation-04-podcast-segment-timed-u06-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 76197, "plain": "That is the idea: control the intention, measure the result, and listen carefully.", "source_duration_s": 4.48, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 11520, "last_word_sample": 207360, "script_start_s": 0.24, "script_end_s": 4.32, "all_asr_last_word_s": 4.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.41470921625353163, "source_start_sample": 7680, "source_end_sample": 214560, "first_word_output_sample": 3840, "last_word_output_sample": 199680, "samples": 206880, "sample_rate": 48000, "head_removed_s": 0.16, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-timed-u06-c00.mp3" }, "en-conversation-01-podcast-segment-free-u03-c00": { "id": "en-conversation-01-podcast-segment-free-u03-c00", "case": "en-conversation-01", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 73149, "plain": "Wait, so I can ask one voice to doubt you while the other celebrates?", "source_duration_s": 8.56, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 218880, "script_start_s": 0.0, "script_end_s": 4.56, "all_asr_last_word_s": 4.5600000000000005 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-01-podcast-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.15054330961556348, "source_start_sample": 0, "source_end_sample": 226080, "first_word_output_sample": 0, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 3.85, "added_silence_s": 0.0, "duration_s": 4.71, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-01-podcast-segment-free-u03-c00.mp3" }, "en-conversation-02-podcast-segment-free-u03-c00": { "id": "en-conversation-02-podcast-segment-free-u03-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 74149, "plain": "And does the second host interrupt with healthy skepticism, or just politely agree?", "source_duration_s": 10.96, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 268800, "script_start_s": 0.16, "script_end_s": 5.6, "all_asr_last_word_s": 10.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-free-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0, "source_identity_reward": 0.2689472111883059, "source_start_sample": 3840, "source_end_sample": 276000, "first_word_output_sample": 3840, "last_word_output_sample": 264960, "samples": 272160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 5.21, "added_silence_s": 0.0, "duration_s": 5.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-free-u03-c00.mp3" }, "de-conversation-02-podcast-segment-timed-u00-c00": { "id": "de-conversation-02-podcast-segment-timed-u00-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 84101, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken.", "source_duration_s": 4.8, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 226560, "script_start_s": 0.08, "script_end_s": 4.72, "all_asr_last_word_s": 4.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7364211348125134, "source_start_sample": 0, "source_end_sample": 230400, "first_word_output_sample": 3840, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-timed-u00-c00.mp3" }, "de-conversation-04-podcast-block-timed-u01-c00": { "id": "de-conversation-04-podcast-block-timed-u01-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 86617, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.2, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 65, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1102080, "script_start_s": 0.0, "script_end_s": 22.96, "all_asr_last_word_s": 22.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6994121583801185, "source_start_sample": 0, "source_end_sample": 1109280, "first_word_output_sample": 0, "last_word_output_sample": 1102080, "samples": 1109280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 23.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u01-c00.mp3" }, "de-conversation-04-podcast-block-timed-u01-c01": { "id": "de-conversation-04-podcast-block-timed-u01-c01", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 86618, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.12, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 65, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1105920, "script_start_s": 0.0, "script_end_s": 23.04, "all_asr_last_word_s": 23.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7487411001568463, "source_start_sample": 0, "source_end_sample": 1109760, "first_word_output_sample": 0, "last_word_output_sample": 1105920, "samples": 1113120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u01-c01.mp3" }, "de-conversation-04-podcast-block-timed-u01-c02": { "id": "de-conversation-04-podcast-block-timed-u01-c02", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 86619, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.12, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 63, "coverage": 0.9692307692307692, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1102080, "script_start_s": 0.0, "script_end_s": 22.96, "all_asr_last_word_s": 22.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03076923076923077, "source_identity_reward": 0.7167691805398474, "source_start_sample": 0, "source_end_sample": 1109280, "first_word_output_sample": 0, "last_word_output_sample": 1102080, "samples": 1109280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u01-c02.mp3" }, "de-conversation-04-podcast-block-timed-u01-c03": { "id": "de-conversation-04-podcast-block-timed-u01-c03", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 86620, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.2, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 61, "coverage": 0.9384615384615385, "first_script_word_verified": false, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06153846153846154, "source_identity_reward": 0.6754536295468264 }, "de-conversation-04-podcast-block-timed-u01-c04": { "id": "de-conversation-04-podcast-block-timed-u01-c04", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 86621, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.12, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 61, "coverage": 0.9384615384615385, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1102080, "script_start_s": 0.0, "script_end_s": 22.96, "all_asr_last_word_s": 22.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06153846153846154, "source_identity_reward": 0.736354940621138, "source_start_sample": 0, "source_end_sample": 1109280, "first_word_output_sample": 0, "last_word_output_sample": 1102080, "samples": 1109280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u01-c04.mp3" }, "de-conversation-04-podcast-block-timed-u01-c05": { "id": "de-conversation-04-podcast-block-timed-u01-c05", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 86622, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.12, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 64, "coverage": 0.9846153846153847, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1105920, "script_start_s": 0.0, "script_end_s": 23.04, "all_asr_last_word_s": 23.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03076923076923077, "source_identity_reward": 0.6815245384847146, "source_start_sample": 0, "source_end_sample": 1109760, "first_word_output_sample": 0, "last_word_output_sample": 1105920, "samples": 1113120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u01-c05.mp3" }, "de-conversation-04-podcast-block-timed-u01-c06": { "id": "de-conversation-04-podcast-block-timed-u01-c06", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 86623, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.2, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 63, "coverage": 0.9692307692307692, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1109760, "script_start_s": 0.0, "script_end_s": 23.12, "all_asr_last_word_s": 23.12 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03076923076923077, "source_identity_reward": 0.6731422352579626, "source_start_sample": 0, "source_end_sample": 1113600, "first_word_output_sample": 0, "last_word_output_sample": 1109760, "samples": 1116960, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.27, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u01-c06.mp3" }, "de-conversation-04-podcast-block-timed-u01-c07": { "id": "de-conversation-04-podcast-block-timed-u01-c07", "case": "de-conversation-04", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 86624, "plain": "Dann lass diesen Satz ruhig klingen, als würde ein Freund nach den richtigen Worten suchen. Genau darum geht es: eine Absicht vorgeben, das Ergebnis messen und aufmerksam zuhören. Ein angenehm menschlicher Umgang mit einer ziemlich eigenartigen kleinen Maschine. Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt. Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 23.2, "source_frame_cap": false, "alignment": { "expected_tokens": 65, "matched_tokens": 63, "coverage": 0.9692307692307692, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1105920, "script_start_s": 0.16, "script_end_s": 23.04, "all_asr_last_word_s": 23.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-block-timed-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03076923076923077, "source_identity_reward": 0.629018269354961, "source_start_sample": 3840, "source_end_sample": 1113120, "first_word_output_sample": 3840, "last_word_output_sample": 1102080, "samples": 1109280, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-block-timed-u01-c07.mp3" }, "de-conversation-03-podcast-segment-free-u05-c00": { "id": "de-conversation-03-podcast-segment-free-u05-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85181, "plain": "Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 337920, "script_start_s": 0.16, "script_end_s": 7.04, "all_asr_last_word_s": 7.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5172949482162159, "source_start_sample": 3840, "source_end_sample": 345120, "first_word_output_sample": 3840, "last_word_output_sample": 334080, "samples": 341280, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 22.81, "added_silence_s": 0.0, "duration_s": 7.11, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-free-u05-c00.mp3" }, "en-conversation-04-podcast-segment-free-u09-c00": { "id": "en-conversation-04-podcast-segment-free-u09-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 30.0, "source_frame_cap": true, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 257280, "script_start_s": 0.16, "script_end_s": 5.36, "all_asr_last_word_s": 19.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-free-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 1.0833333333333333, "source_identity_reward": 0.17335744083144342, "source_start_sample": 3840, "source_end_sample": 264480, "first_word_output_sample": 3840, "last_word_output_sample": 253440, "samples": 260640, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 24.49, "added_silence_s": 0.0, "duration_s": 5.43, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-free-u09-c00.mp3" }, "en-conversation-04-podcast-segment-free-u07-c00": { "id": "en-conversation-04-podcast-segment-free-u07-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 76213, "plain": "A reassuringly human approach to a very peculiar little machine.", "source_duration_s": 7.04, "source_frame_cap": false, "alignment": { "expected_tokens": 10, "matched_tokens": 10, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 334080, "script_start_s": 0.16, "script_end_s": 6.96, "all_asr_last_word_s": 6.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-segment-free-u07-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.14766160466821765, "source_start_sample": 3840, "source_end_sample": 337920, "first_word_output_sample": 3840, "last_word_output_sample": 330240, "samples": 337440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 7.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-segment-free-u07-c00.mp3" }, "en-conversation-05-podcast-block-timed-u03-c00": { "id": "en-conversation-05-podcast-block-timed-u03-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 77649, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 23.04, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1098240, "script_start_s": 0.0, "script_end_s": 22.88, "all_asr_last_word_s": 22.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.6484456241697891, "source_start_sample": 0, "source_end_sample": 1105440, "first_word_output_sample": 0, "last_word_output_sample": 1098240, "samples": 1105440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u03-c00.mp3" }, "en-conversation-05-podcast-block-timed-u03-c01": { "id": "en-conversation-05-podcast-block-timed-u03-c01", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 77650, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 23.04, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1098240, "script_start_s": 0.0, "script_end_s": 22.88, "all_asr_last_word_s": 22.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u03-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.5955558889435634, "source_start_sample": 0, "source_end_sample": 1105440, "first_word_output_sample": 0, "last_word_output_sample": 1098240, "samples": 1105440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u03-c01.mp3" }, "en-conversation-05-podcast-block-timed-u03-c02": { "id": "en-conversation-05-podcast-block-timed-u03-c02", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 77651, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 23.04, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1098240, "script_start_s": 0.16, "script_end_s": 22.88, "all_asr_last_word_s": 22.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u03-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.6005597933087423, "source_start_sample": 3840, "source_end_sample": 1105440, "first_word_output_sample": 3840, "last_word_output_sample": 1094400, "samples": 1101600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u03-c02.mp3" }, "en-conversation-05-podcast-block-timed-u03-c03": { "id": "en-conversation-05-podcast-block-timed-u03-c03", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 77652, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 23.04, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1102080, "script_start_s": 0.0, "script_end_s": 22.96, "all_asr_last_word_s": 22.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u03-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.5415990474030116, "source_start_sample": 0, "source_end_sample": 1105920, "first_word_output_sample": 0, "last_word_output_sample": 1102080, "samples": 1109280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u03-c03.mp3" }, "en-conversation-05-podcast-block-timed-u03-c04": { "id": "en-conversation-05-podcast-block-timed-u03-c04", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 77653, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 22.96, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1098240, "script_start_s": 0.0, "script_end_s": 22.88, "all_asr_last_word_s": 22.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u03-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.6440794212884815, "source_start_sample": 0, "source_end_sample": 1102080, "first_word_output_sample": 0, "last_word_output_sample": 1098240, "samples": 1105440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u03-c04.mp3" }, "en-conversation-05-podcast-block-timed-u03-c05": { "id": "en-conversation-05-podcast-block-timed-u03-c05", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 77654, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 22.96, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 67, "coverage": 0.9710144927536232, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1098240, "script_start_s": 0.0, "script_end_s": 22.88, "all_asr_last_word_s": 22.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u03-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.028985507246376812, "source_identity_reward": 0.568099217340735, "source_start_sample": 0, "source_end_sample": 1102080, "first_word_output_sample": 0, "last_word_output_sample": 1098240, "samples": 1105440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 23.03, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u03-c05.mp3" }, "en-conversation-05-podcast-block-timed-u03-c06": { "id": "en-conversation-05-podcast-block-timed-u03-c06", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 77655, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 23.04, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1098240, "script_start_s": 0.16, "script_end_s": 22.88, "all_asr_last_word_s": 22.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u03-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.589915650285529, "source_start_sample": 3840, "source_end_sample": 1105440, "first_word_output_sample": 3840, "last_word_output_sample": 1094400, "samples": 1101600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u03-c06.mp3" }, "en-conversation-05-podcast-block-timed-u03-c07": { "id": "en-conversation-05-podcast-block-timed-u03-c07", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 77656, "plain": "Imagine using this for a little science programme, where curiosity matters more than grand claims. And the second host is allowed to say, hold on, explain that again, I have not followed you. Exactly. The exchange should feel responsive, with the same two people returning after every change of mood. Then we have a fair test. Keep the words fixed and let the different generation methods speak for themselves.", "source_duration_s": 23.04, "source_frame_cap": false, "alignment": { "expected_tokens": 69, "matched_tokens": 68, "coverage": 0.9855072463768116, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1098240, "script_start_s": 0.16, "script_end_s": 22.88, "all_asr_last_word_s": 22.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u03-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.014492753623188406, "source_identity_reward": 0.570760322516697, "source_start_sample": 3840, "source_end_sample": 1105440, "first_word_output_sample": 3840, "last_word_output_sample": 1094400, "samples": 1101600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u03-c07.mp3" }, "en-conversation-05-podcast-segment-free-u02-c00": { "id": "en-conversation-05-podcast-segment-free-u02-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77133, "plain": "That is what this five-language grid explores, with short and longer conversations.", "source_duration_s": 5.52, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 249600, "script_start_s": 0.0, "script_end_s": 5.2, "all_asr_last_word_s": 5.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.44059881282267055, "source_start_sample": 0, "source_end_sample": 256800, "first_word_output_sample": 0, "last_word_output_sample": 249600, "samples": 256800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.17, "added_silence_s": 0.0, "duration_s": 5.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-free-u02-c00.mp3" }, "de-conversation-03-podcast-segment-free-u08-c00": { "id": "de-conversation-03-podcast-segment-free-u08-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 85229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 7.28, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 299520, "script_start_s": 0.0, "script_end_s": 6.24, "all_asr_last_word_s": 6.24 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6868146803265518, "source_start_sample": 0, "source_end_sample": 306720, "first_word_output_sample": 0, "last_word_output_sample": 299520, "samples": 306720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.89, "added_silence_s": 0.0, "duration_s": 6.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-free-u08-c00.mp3" }, "en-conversation-05-podcast-block-timed-u02-c00": { "id": "en-conversation-05-podcast-block-timed-u02-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 0, "seed": 77633, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 871680, "script_start_s": 0.16, "script_end_s": 18.16, "all_asr_last_word_s": 18.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4928230101478632, "source_start_sample": 3840, "source_end_sample": 878880, "first_word_output_sample": 3840, "last_word_output_sample": 867840, "samples": 875040, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 18.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u02-c00.mp3" }, "en-conversation-05-podcast-block-timed-u02-c01": { "id": "en-conversation-05-podcast-block-timed-u02-c01", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 1, "seed": 77634, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 871680, "script_start_s": 0.0, "script_end_s": 18.16, "all_asr_last_word_s": 18.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.4872453757399311, "source_start_sample": 0, "source_end_sample": 878880, "first_word_output_sample": 0, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 18.31, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u02-c01.mp3" }, "en-conversation-05-podcast-block-timed-u02-c02": { "id": "en-conversation-05-podcast-block-timed-u02-c02", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 2, "seed": 77635, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 875520, "script_start_s": 0.16, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03636363636363636, "source_identity_reward": 0.4700247707442907, "source_start_sample": 3840, "source_end_sample": 879360, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u02-c02.mp3" }, "en-conversation-05-podcast-block-timed-u02-c03": { "id": "en-conversation-05-podcast-block-timed-u02-c03", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 3, "seed": 77636, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.4, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 879360, "script_start_s": 0.0, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.517395302560463, "source_start_sample": 0, "source_end_sample": 883200, "first_word_output_sample": 0, "last_word_output_sample": 879360, "samples": 886560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u02-c03.mp3" }, "en-conversation-05-podcast-block-timed-u02-c04": { "id": "en-conversation-05-podcast-block-timed-u02-c04", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 4, "seed": 77637, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 875520, "script_start_s": 0.16, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.3916503107574606, "source_start_sample": 3840, "source_end_sample": 879360, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u02-c04.mp3" }, "en-conversation-05-podcast-block-timed-u02-c05": { "id": "en-conversation-05-podcast-block-timed-u02-c05", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 5, "seed": 77638, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 875520, "script_start_s": 0.0, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.03636363636363636, "source_identity_reward": 0.4290565126046024, "source_start_sample": 0, "source_end_sample": 879360, "first_word_output_sample": 0, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.39, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u02-c05.mp3" }, "en-conversation-05-podcast-block-timed-u02-c06": { "id": "en-conversation-05-podcast-block-timed-u02-c06", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 6, "seed": 77639, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 875520, "script_start_s": 0.16, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5880579837904855, "source_start_sample": 3840, "source_end_sample": 879360, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u02-c06.mp3" }, "en-conversation-05-podcast-block-timed-u02-c07": { "id": "en-conversation-05-podcast-block-timed-u02-c07", "case": "en-conversation-05", "model": "podcast", "strategy": "block", "timing": "timed", "candidate": 7, "seed": 77640, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.32, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 875520, "script_start_s": 0.16, "script_end_s": 18.24, "all_asr_last_word_s": 18.240000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-block-timed-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.4894845345135772, "source_start_sample": 3840, "source_end_sample": 879360, "first_word_output_sample": 3840, "last_word_output_sample": 871680, "samples": 878880, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.31, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-block-timed-u02-c07.mp3" }, "en-conversation-02-podcast-segment-timed-u09-c00": { "id": "en-conversation-02-podcast-segment-timed-u09-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 192000, "script_start_s": 0.16, "script_end_s": 4.0, "all_asr_last_word_s": 4.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2674307608845617, "source_start_sample": 3840, "source_end_sample": 195840, "first_word_output_sample": 3840, "last_word_output_sample": 188160, "samples": 195360, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.07, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-timed-u09-c00.mp3" }, "en-conversation-04-podcast-block-free-u01-c00": { "id": "en-conversation-04-podcast-block-free-u01-c00", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 76617, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 31.44, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 62, "coverage": 0.9841269841269841, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1094400, "script_start_s": 0.16, "script_end_s": 22.8, "all_asr_last_word_s": 28.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07936507936507936, "source_identity_reward": 0.4007804842361961, "source_start_sample": 3840, "source_end_sample": 1101600, "first_word_output_sample": 3840, "last_word_output_sample": 1090560, "samples": 1097760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 8.49, "added_silence_s": 0.0, "duration_s": 22.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u01-c00.mp3" }, "en-conversation-04-podcast-block-free-u01-c01": { "id": "en-conversation-04-podcast-block-free-u01-c01", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 76618, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 22.64, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1079040, "script_start_s": 0.16, "script_end_s": 22.48, "all_asr_last_word_s": 22.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5831531093701245, "source_start_sample": 3840, "source_end_sample": 1086240, "first_word_output_sample": 3840, "last_word_output_sample": 1075200, "samples": 1082400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u01-c01.mp3" }, "en-conversation-04-podcast-block-free-u01-c02": { "id": "en-conversation-04-podcast-block-free-u01-c02", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 76619, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 22.24, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 63, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1059840, "script_start_s": 0.08, "script_end_s": 22.08, "all_asr_last_word_s": 22.080000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4570028711084695, "source_start_sample": 0, "source_end_sample": 1067040, "first_word_output_sample": 3840, "last_word_output_sample": 1059840, "samples": 1067040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 22.23, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u01-c02.mp3" }, "en-conversation-04-podcast-block-free-u01-c03": { "id": "en-conversation-04-podcast-block-free-u01-c03", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 76620, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 22.0, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 62, "coverage": 0.9841269841269841, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1048320, "script_start_s": 0.08, "script_end_s": 21.84, "all_asr_last_word_s": 21.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.047619047619047616, "source_identity_reward": 0.4705194488368769, "source_start_sample": 0, "source_end_sample": 1055520, "first_word_output_sample": 3840, "last_word_output_sample": 1048320, "samples": 1055520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.99, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u01-c03.mp3" }, "en-conversation-04-podcast-block-free-u01-c04": { "id": "en-conversation-04-podcast-block-free-u01-c04", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 76621, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 61, "coverage": 0.9682539682539683, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1036800, "script_start_s": 0.16, "script_end_s": 21.6, "all_asr_last_word_s": 21.6 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.047619047619047616, "source_identity_reward": 0.5383666340486924, "source_start_sample": 3840, "source_end_sample": 1044000, "first_word_output_sample": 3840, "last_word_output_sample": 1032960, "samples": 1040160, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 21.67, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u01-c04.mp3" }, "en-conversation-04-podcast-block-free-u01-c05": { "id": "en-conversation-04-podcast-block-free-u01-c05", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 76622, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 36.72, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 61, "coverage": 0.9682539682539683, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1086720, "script_start_s": 0.16, "script_end_s": 22.64, "all_asr_last_word_s": 36.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.31746031746031744, "source_identity_reward": 0.43733640292943027, "source_start_sample": 3840, "source_end_sample": 1093920, "first_word_output_sample": 3840, "last_word_output_sample": 1082880, "samples": 1090080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 13.93, "added_silence_s": 0.0, "duration_s": 22.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u01-c05.mp3" }, "en-conversation-04-podcast-block-free-u01-c06": { "id": "en-conversation-04-podcast-block-free-u01-c06", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 76623, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 23.04, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 62, "coverage": 0.9841269841269841, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 3840, "last_word_sample": 1098240, "script_start_s": 0.08, "script_end_s": 22.88, "all_asr_last_word_s": 22.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.031746031746031744, "source_identity_reward": 0.5018719908349196, "source_start_sample": 0, "source_end_sample": 1105440, "first_word_output_sample": 3840, "last_word_output_sample": 1098240, "samples": 1105440, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 23.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u01-c06.mp3" }, "en-conversation-04-podcast-block-free-u01-c07": { "id": "en-conversation-04-podcast-block-free-u01-c07", "case": "en-conversation-04", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 76624, "plain": "Then make this sentence unhurried, as if a friend were choosing the right words. That is the idea: control the intention, measure the result, and listen carefully. A reassuringly human approach to a very peculiar little machine. There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 32.88, "source_frame_cap": false, "alignment": { "expected_tokens": 63, "matched_tokens": 62, "coverage": 0.9841269841269841, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 1075200, "script_start_s": 0.16, "script_end_s": 22.4, "all_asr_last_word_s": 32.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-04-podcast-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3492063492063492, "source_identity_reward": 0.31360897079378625, "source_start_sample": 3840, "source_end_sample": 1082400, "first_word_output_sample": 3840, "last_word_output_sample": 1071360, "samples": 1078560, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 10.33, "added_silence_s": 0.0, "duration_s": 22.47, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-04-podcast-block-free-u01-c07.mp3" }, "de-conversation-05-podcast-segment-free-u05-c00": { "id": "de-conversation-05-podcast-segment-free-u05-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 87181, "plain": "Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen.", "source_duration_s": 10.32, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 376320, "script_start_s": 0.0, "script_end_s": 7.84, "all_asr_last_word_s": 7.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-free-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4242561167640258, "source_start_sample": 0, "source_end_sample": 383520, "first_word_output_sample": 0, "last_word_output_sample": 376320, "samples": 383520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 2.33, "added_silence_s": 0.0, "duration_s": 7.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-free-u05-c00.mp3" }, "de-conversation-04-podcast-segment-timed-u04-c00": { "id": "de-conversation-04-podcast-segment-timed-u04-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86165, "plain": "Wir messen die tatsächliche Dauer und zeigen auch vorzeitige Enden und erreichte Längenlimits.", "source_duration_s": 4.8, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 12, "coverage": 0.9230769230769231, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 222720, "script_start_s": 0.0, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07692307692307693, "source_identity_reward": 0.6341089197349976, "source_start_sample": 0, "source_end_sample": 229920, "first_word_output_sample": 0, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-timed-u04-c00.mp3" }, "de-conversation-04-podcast-segment-free-u02-c00": { "id": "de-conversation-04-podcast-segment-free-u02-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 86133, "plain": "Nein. Der Codec arbeitet in Achtzig-Millisekunden-Schritten, und die gewünschte Länge kann verfehlt werden.", "source_duration_s": 7.44, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 11, "coverage": 0.7333333333333333, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 318720, "script_start_s": 0.0, "script_end_s": 6.64, "all_asr_last_word_s": 6.640000000000001 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.26666666666666666, "source_identity_reward": 0.5610264985153193, "source_start_sample": 0, "source_end_sample": 325920, "first_word_output_sample": 0, "last_word_output_sample": 318720, "samples": 325920, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.65, "added_silence_s": 0.0, "duration_s": 6.79, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-free-u02-c00.mp3" }, "de-conversation-05-podcast-segment-timed-u03-c00": { "id": "de-conversation-05-podcast-segment-timed-u03-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87149, "plain": "Ich würde ohne Referenz anfangen und zwei Gastgeber erfinden lassen. Gewagt, aber spannend.", "source_duration_s": 4.8, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 12, "coverage": 0.9230769230769231, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 226560, "script_start_s": 0.16, "script_end_s": 4.72, "all_asr_last_word_s": 4.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u03-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.07692307692307693, "source_identity_reward": 0.6043995587837052, "source_start_sample": 3840, "source_end_sample": 230400, "first_word_output_sample": 3840, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-timed-u03-c00.mp3" }, "de-conversation-04-podcast-segment-timed-u09-c00": { "id": "de-conversation-04-podcast-segment-timed-u09-c00", "case": "de-conversation-04", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 86245, "plain": "Ja, manchmal erzählt dieses kleine Zögern mehr als der nächste völlig glatt formulierte Satz.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 238080, "script_start_s": 0.0, "script_end_s": 4.96, "all_asr_last_word_s": 4.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-04-podcast-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4655385776252462, "source_start_sample": 0, "source_end_sample": 241920, "first_word_output_sample": 0, "last_word_output_sample": 238080, "samples": 245280, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 5.11, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-04-podcast-segment-timed-u09-c00.mp3" }, "de-conversation-03-podcast-segment-timed-u08-c00": { "id": "de-conversation-03-podcast-segment-timed-u08-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 85229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 4.8, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 226560, "script_start_s": 0.0, "script_end_s": 4.72, "all_asr_last_word_s": 4.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4797967063163413, "source_start_sample": 0, "source_end_sample": 230400, "first_word_output_sample": 0, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-segment-timed-u08-c00.mp3" }, "de-conversation-05-podcast-segment-timed-u08-c00": { "id": "de-conversation-05-podcast-segment-timed-u08-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87229, "plain": "Ich mag Gespräche, in denen noch Platz für eine echte, unerwartete Reaktion bleibt.", "source_duration_s": 4.8, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 226560, "script_start_s": 0.0, "script_end_s": 4.72, "all_asr_last_word_s": 4.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7390052955181683, "source_start_sample": 0, "source_end_sample": 230400, "first_word_output_sample": 0, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-05-podcast-segment-timed-u08-c00.mp3" }, "de-conversation-05-podcast-segment-timed-u05-c00": { "id": "de-conversation-05-podcast-segment-timed-u05-c00", "case": "de-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 87181, "plain": "Ich wünsche mir Wärme, Überraschung und ein bisschen Neckerei, aber keine vorgelesenen Regieanweisungen.", "source_duration_s": 4.8, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 11, "coverage": 0.8461538461538461, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-05-podcast-segment-timed-u05-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.15384615384615385, "source_identity_reward": 0.23203941720367016 }, "en-conversation-05-podcast-segment-timed-u09-c00": { "id": "en-conversation-05-podcast-segment-timed-u09-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 77245, "plain": "Yes, a tiny hesitation can say more than another perfectly polished sentence.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 12, "matched_tokens": 12, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 192000, "script_start_s": 0.0, "script_end_s": 4.0, "all_asr_last_word_s": 4.0 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-timed-u09-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.2145203392637523, "source_start_sample": 0, "source_end_sample": 195840, "first_word_output_sample": 0, "last_word_output_sample": 192000, "samples": 199200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-timed-u09-c00.mp3" }, "de-conversation-01-podcast-segment-timed-u11-c00": { "id": "de-conversation-01-podcast-segment-timed-u11-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83277, "plain": "Einverstanden. Ich möchte Wärme und Persönlichkeit hören und trotzdem erkennen, wer gerade spricht.", "source_duration_s": 4.8, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 226560, "script_start_s": 0.16, "script_end_s": 4.72, "all_asr_last_word_s": 4.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.5731963743497515, "source_start_sample": 3840, "source_end_sample": 230400, "first_word_output_sample": 3840, "last_word_output_sample": 222720, "samples": 229920, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.79, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-timed-u11-c00.mp3" }, "en-conversation-03-podcast-block-free-u02-c00": { "id": "en-conversation-03-podcast-block-free-u02-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 75633, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 25.52, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 910080, "script_start_s": 0.16, "script_end_s": 18.96, "all_asr_last_word_s": 23.44 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u02-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.14545454545454545, "source_identity_reward": 0.35586631863415563, "source_start_sample": 3840, "source_end_sample": 917280, "first_word_output_sample": 3840, "last_word_output_sample": 906240, "samples": 913440, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 6.41, "added_silence_s": 0.0, "duration_s": 19.03, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u02-c00.mp3" }, "en-conversation-03-podcast-block-free-u02-c01": { "id": "en-conversation-03-podcast-block-free-u02-c01", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 75634, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.48, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 879360, "script_start_s": 0.16, "script_end_s": 18.32, "all_asr_last_word_s": 18.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u02-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.39642970981401604, "source_start_sample": 3840, "source_end_sample": 886560, "first_word_output_sample": 3840, "last_word_output_sample": 875520, "samples": 882720, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 18.39, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u02-c01.mp3" }, "en-conversation-03-podcast-block-free-u02-c02": { "id": "en-conversation-03-podcast-block-free-u02-c02", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 75635, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.24, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 54, "coverage": 0.9818181818181818, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 867840, "script_start_s": 0.16, "script_end_s": 18.08, "all_asr_last_word_s": 18.080000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u02-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.01818181818181818, "source_identity_reward": 0.5051274181245157, "source_start_sample": 3840, "source_end_sample": 875040, "first_word_output_sample": 3840, "last_word_output_sample": 864000, "samples": 871200, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 18.15, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u02-c02.mp3" }, "en-conversation-03-podcast-block-free-u02-c03": { "id": "en-conversation-03-podcast-block-free-u02-c03", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 75636, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 20.24, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 956160, "script_start_s": 0.0, "script_end_s": 19.92, "all_asr_last_word_s": 19.92 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u02-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.40892817020622574, "source_start_sample": 0, "source_end_sample": 963360, "first_word_output_sample": 0, "last_word_output_sample": 956160, "samples": 963360, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.17, "added_silence_s": 0.0, "duration_s": 20.07, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u02-c03.mp3" }, "en-conversation-03-podcast-block-free-u02-c04": { "id": "en-conversation-03-podcast-block-free-u02-c04", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 75637, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.72, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 894720, "script_start_s": 0.16, "script_end_s": 18.64, "all_asr_last_word_s": 18.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u02-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.46982986784983655, "source_start_sample": 3840, "source_end_sample": 898560, "first_word_output_sample": 3840, "last_word_output_sample": 890880, "samples": 898080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u02-c04.mp3" }, "en-conversation-03-podcast-block-free-u02-c05": { "id": "en-conversation-03-podcast-block-free-u02-c05", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 75638, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.72, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 887040, "script_start_s": 0.16, "script_end_s": 18.48, "all_asr_last_word_s": 18.48 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u02-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.43250609157813935, "source_start_sample": 3840, "source_end_sample": 894240, "first_word_output_sample": 3840, "last_word_output_sample": 883200, "samples": 890400, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.09, "added_silence_s": 0.0, "duration_s": 18.55, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u02-c05.mp3" }, "en-conversation-03-podcast-block-free-u02-c06": { "id": "en-conversation-03-podcast-block-free-u02-c06", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 75639, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 18.88, "source_frame_cap": false, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 902400, "script_start_s": 0.16, "script_end_s": 18.8, "all_asr_last_word_s": 18.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u02-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.4475163936421571, "source_start_sample": 3840, "source_end_sample": 906240, "first_word_output_sample": 3840, "last_word_output_sample": 898560, "samples": 905760, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 18.87, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u02-c06.mp3" }, "en-conversation-03-podcast-block-free-u02-c07": { "id": "en-conversation-03-podcast-block-free-u02-c07", "case": "en-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 75640, "plain": "There is something lovely about a conversation that leaves room for a real reaction. Yes, a tiny hesitation can say more than another perfectly polished sentence. So let us compare the voices carefully, including the moments where the illusion breaks. Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 60.0, "source_frame_cap": true, "alignment": { "expected_tokens": 55, "matched_tokens": 55, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 852480, "script_start_s": 0.16, "script_end_s": 17.76, "all_asr_last_word_s": 59.8 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-block-free-u02-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2727272727272727, "source_identity_reward": 0.4291305560547089, "source_start_sample": 3840, "source_end_sample": 859680, "first_word_output_sample": 3840, "last_word_output_sample": 848640, "samples": 855840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 42.09, "added_silence_s": 0.0, "duration_s": 17.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-block-free-u02-c07.mp3" }, "de-conversation-01-podcast-segment-timed-u06-c00": { "id": "de-conversation-01-podcast-segment-timed-u06-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83197, "plain": "Im Training kommen echte Gespräche und Schauspielbeispiele zusammen, damit beide Formate erhalten bleiben.", "source_duration_s": 4.8, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 226560, "script_start_s": 0.0, "script_end_s": 4.72, "all_asr_last_word_s": 4.72 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-timed-u06-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7364916088470732, "source_start_sample": 0, "source_end_sample": 230400, "first_word_output_sample": 0, "last_word_output_sample": 226560, "samples": 233760, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.87, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-01-podcast-segment-timed-u06-c00.mp3" }, "en-conversation-02-podcast-segment-timed-u11-c00": { "id": "en-conversation-02-podcast-segment-timed-u11-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74277, "plain": "Agreed. I want warmth and character, and I still want to recognize who is talking.", "source_duration_s": 5.04, "source_frame_cap": false, "alignment": { "expected_tokens": 15, "matched_tokens": 15, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 234240, "script_start_s": 0.16, "script_end_s": 4.88, "all_asr_last_word_s": 4.88 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-timed-u11-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.31902387015882455, "source_start_sample": 3840, "source_end_sample": 241440, "first_word_output_sample": 3840, "last_word_output_sample": 230400, "samples": 237600, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 4.95, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-timed-u11-c00.mp3" }, "en-conversation-03-podcast-segment-free-u08-c00": { "id": "en-conversation-03-podcast-segment-free-u08-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 75229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 7.6, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 272640, "script_start_s": 0.16, "script_end_s": 5.68, "all_asr_last_word_s": 5.68 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-free-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.19040084299706733, "source_start_sample": 3840, "source_end_sample": 279840, "first_word_output_sample": 3840, "last_word_output_sample": 268800, "samples": 276000, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 1.77, "added_silence_s": 0.0, "duration_s": 5.75, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-free-u08-c00.mp3" }, "de-conversation-02-podcast-segment-free-u00-c00": { "id": "de-conversation-02-podcast-segment-free-u00-c00", "case": "de-conversation-02", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 84101, "plain": "Stell dir ein winziges Radiostudio vor, in dem zwei Leute dieses Modell entdecken.", "source_duration_s": 11.28, "source_frame_cap": false, "alignment": { "expected_tokens": 13, "matched_tokens": 13, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 280320, "script_start_s": 0.0, "script_end_s": 5.84, "all_asr_last_word_s": 5.84 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-02-podcast-segment-free-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.6723571613370369, "source_start_sample": 0, "source_end_sample": 287520, "first_word_output_sample": 0, "last_word_output_sample": 280320, "samples": 287520, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 5.29, "added_silence_s": 0.0, "duration_s": 5.99, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-02-podcast-segment-free-u00-c00.mp3" }, "de-conversation-03-podcast-block-free-u01-c00": { "id": "de-conversation-03-podcast-block-free-u01-c00", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 0, "seed": 85617, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 20.48, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 47, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 975360, "script_start_s": 0.0, "script_end_s": 20.32, "all_asr_last_word_s": 20.32 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u01-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.8698175204574583, "source_start_sample": 0, "source_end_sample": 982560, "first_word_output_sample": 0, "last_word_output_sample": 975360, "samples": 982560, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.01, "added_silence_s": 0.0, "duration_s": 20.47, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u01-c00.mp3" }, "de-conversation-03-podcast-block-free-u01-c01": { "id": "de-conversation-03-podcast-block-free-u01-c01", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 1, "seed": 85618, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 21.76, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 47, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1017600, "script_start_s": 0.0, "script_end_s": 21.2, "all_asr_last_word_s": 21.52 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u01-c01.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.02127659574468085, "source_identity_reward": 0.6519948890968316, "source_start_sample": 0, "source_end_sample": 1024800, "first_word_output_sample": 0, "last_word_output_sample": 1017600, "samples": 1024800, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.41, "added_silence_s": 0.0, "duration_s": 21.35, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u01-c01.mp3" }, "de-conversation-03-podcast-block-free-u01-c02": { "id": "de-conversation-03-podcast-block-free-u01-c02", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 2, "seed": 85619, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 22.72, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 47, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1059840, "script_start_s": 0.0, "script_end_s": 22.08, "all_asr_last_word_s": 22.080000000000002 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u01-c02.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.7740817058393687, "source_start_sample": 0, "source_end_sample": 1067040, "first_word_output_sample": 0, "last_word_output_sample": 1059840, "samples": 1067040, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 0.49, "added_silence_s": 0.0, "duration_s": 22.23, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u01-c02.mp3" }, "de-conversation-03-podcast-block-free-u01-c03": { "id": "de-conversation-03-podcast-block-free-u01-c03", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 3, "seed": 85620, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 35.76, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 45, "coverage": 0.9574468085106383, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1032960, "script_start_s": 0.0, "script_end_s": 21.52, "all_asr_last_word_s": 35.519999999999996 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u01-c03.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.3191489361702128, "source_identity_reward": 0.6272095369793818, "source_start_sample": 0, "source_end_sample": 1040160, "first_word_output_sample": 0, "last_word_output_sample": 1032960, "samples": 1040160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 14.09, "added_silence_s": 0.0, "duration_s": 21.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u01-c03.mp3" }, "de-conversation-03-podcast-block-free-u01-c04": { "id": "de-conversation-03-podcast-block-free-u01-c04", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 4, "seed": 85621, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 25.76, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 47, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1032960, "script_start_s": 0.0, "script_end_s": 21.52, "all_asr_last_word_s": 24.96 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u01-c04.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.06382978723404255, "source_identity_reward": 0.7467939618052835, "source_start_sample": 0, "source_end_sample": 1040160, "first_word_output_sample": 0, "last_word_output_sample": 1032960, "samples": 1040160, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 4.09, "added_silence_s": 0.0, "duration_s": 21.67, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u01-c04.mp3" }, "de-conversation-03-podcast-block-free-u01-c05": { "id": "de-conversation-03-podcast-block-free-u01-c05", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 5, "seed": 85622, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 30.16, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 46, "coverage": 0.9787234042553191, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1029120, "script_start_s": 0.0, "script_end_s": 21.44, "all_asr_last_word_s": 30.04 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u01-c05.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2765957446808511, "source_identity_reward": 0.5358036138287083, "source_start_sample": 0, "source_end_sample": 1036320, "first_word_output_sample": 0, "last_word_output_sample": 1029120, "samples": 1036320, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 8.57, "added_silence_s": 0.0, "duration_s": 21.59, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u01-c05.mp3" }, "de-conversation-03-podcast-block-free-u01-c06": { "id": "de-conversation-03-podcast-block-free-u01-c06", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 6, "seed": 85623, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 22.32, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 47, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1009920, "script_start_s": 0.0, "script_end_s": 21.04, "all_asr_last_word_s": 22.16 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u01-c06.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0425531914893617, "source_identity_reward": 0.7151294233677794, "source_start_sample": 0, "source_end_sample": 1017120, "first_word_output_sample": 0, "last_word_output_sample": 1009920, "samples": 1017120, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 1.13, "added_silence_s": 0.0, "duration_s": 21.19, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u01-c06.mp3" }, "de-conversation-03-podcast-block-free-u01-c07": { "id": "de-conversation-03-podcast-block-free-u01-c07", "case": "de-conversation-03", "model": "podcast", "strategy": "block", "timing": "free", "candidate": 7, "seed": 85624, "plain": "Darum vergleichen wir die einzelnen Turns und vertrauen nicht einfach dem ersten Satz. Gut. Die begeisterte Gastgeberin soll dieselbe Person bleiben, wenn sie plötzlich zweifelt. Referenzen steuern die Identität, Spielanweisungen den Ausdruck. Beides ist keine Zaubergarantie. Seltsam, die ehrliche Erklärung macht mich neugieriger als das große Werbeversprechen.", "source_duration_s": 32.88, "source_frame_cap": false, "alignment": { "expected_tokens": 47, "matched_tokens": 46, "coverage": 0.9787234042553191, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 0, "last_word_sample": 1056000, "script_start_s": 0.0, "script_end_s": 22.0, "all_asr_last_word_s": 32.2 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-03-podcast-block-free-u01-c07.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.2553191489361702, "source_identity_reward": 0.6603929993471558, "source_start_sample": 0, "source_end_sample": 1063200, "first_word_output_sample": 0, "last_word_output_sample": 1056000, "samples": 1063200, "sample_rate": 48000, "head_removed_s": 0.0, "tail_removed_s": 10.73, "added_silence_s": 0.0, "duration_s": 22.15, "lead_ms": 0.0, "tail_ms": 150.0, "fade_in_ms": 0.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/de-conversation-03-podcast-block-free-u01-c07.mp3" }, "en-conversation-02-podcast-segment-timed-u04-c00": { "id": "en-conversation-02-podcast-segment-timed-u04-c00", "case": "en-conversation-02", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 74165, "plain": "Healthy skepticism, please. We give each turn its own mood and its own words.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 222720, "script_start_s": 0.16, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-02-podcast-segment-timed-u04-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.23287940452658373, "source_start_sample": 3840, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-02-podcast-segment-timed-u04-c00.mp3" }, "de-conversation-01-podcast-segment-timed-u00-c00": { "id": "de-conversation-01-podcast-segment-timed-u00-c00", "case": "de-conversation-01", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 83101, "plain": "Ich habe Humaneness Podcast gehört, eine Feinabstimmung von Humaneness Voice Large.", "source_duration_s": 4.08, "source_frame_cap": false, "alignment": { "expected_tokens": 11, "matched_tokens": 8, "coverage": 0.7272727272727273, "first_script_word_verified": true, "last_script_word_verified": false, "all_intended_turns_aligned": true, "ready": false, "reason": "Script endpoint or an intended turn was not reliably recognized; raw take retained for review. No guessed cut." }, "ready": false, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/de-conversation-01-podcast-segment-timed-u00-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.36363636363636365, "source_identity_reward": 0.47718369674206945 }, "en-conversation-05-podcast-segment-free-u13-c00": { "id": "en-conversation-05-podcast-segment-free-u13-c00", "case": "en-conversation-05", "model": "podcast", "strategy": "segment", "timing": "free", "candidate": 0, "seed": 77309, "plain": "And the second host is allowed to say, hold on, explain that again, I have not followed you.", "source_duration_s": 8.72, "source_frame_cap": false, "alignment": { "expected_tokens": 18, "matched_tokens": 18, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 276480, "script_start_s": 0.16, "script_end_s": 5.76, "all_asr_last_word_s": 5.76 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-05-podcast-segment-free-u13-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.28464978502652205, "source_start_sample": 3840, "source_end_sample": 283680, "first_word_output_sample": 3840, "last_word_output_sample": 272640, "samples": 279840, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 2.81, "added_silence_s": 0.0, "duration_s": 5.83, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-05-podcast-segment-free-u13-c00.mp3" }, "en-conversation-03-podcast-segment-timed-u08-c00": { "id": "en-conversation-03-podcast-segment-timed-u08-c00", "case": "en-conversation-03", "model": "podcast", "strategy": "segment", "timing": "timed", "candidate": 0, "seed": 75229, "plain": "There is something lovely about a conversation that leaves room for a real reaction.", "source_duration_s": 4.72, "source_frame_cap": false, "alignment": { "expected_tokens": 14, "matched_tokens": 14, "coverage": 1.0, "first_script_word_verified": true, "last_script_word_verified": true, "all_intended_turns_aligned": true, "ready": true, "first_word_sample": 7680, "last_word_sample": 222720, "script_start_s": 0.16, "script_end_s": 4.64, "all_asr_last_word_s": 4.64 }, "ready": true, "raw_audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/raw/en-conversation-03-podcast-segment-timed-u08-c00.mp3", "metrics_scope": "Original untrimmed raw audio only; edited audio has not been rescored.", "source_wer": 0.0, "source_identity_reward": 0.30020423571709587, "source_start_sample": 3840, "source_end_sample": 226560, "first_word_output_sample": 3840, "last_word_output_sample": 218880, "samples": 226080, "sample_rate": 48000, "head_removed_s": 0.08, "tail_removed_s": 0.0, "added_silence_s": 0.07, "duration_s": 4.71, "lead_ms": 80.0, "tail_ms": 150.0, "fade_in_ms": 20.0, "fade_out_ms": 50.0, "audio": "https://huggingface.co/datasets/laion/Humaneness-Podcast-Demo-Audio/resolve/main/segment_block_pilot_20261107/editing_v1/clips/en-conversation-03-podcast-segment-timed-u08-c00.mp3" } }