File size: 4,286 Bytes
383c94a 9fb3018 383c94a 9fb3018 383c94a 9fb3018 383c94a 9fb3018 383c94a | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 | {
"rows": 313655,
"hours": 2055.0831158610463,
"language": "ru",
"previous_language_music_filter": {
"source_repo": "TeraTTS/webinars",
"source_revision": "83c2459778926a7db6144abd552b084de1ac95e9",
"input_rows": 327254,
"input_hours": 2139.101691971096,
"rows": 315085,
"hours": 2064.612951877606,
"seconds": 7432606.626759382,
"removed": {
"non_russian_language_tag": 3526,
"detected_music_or_singing": 8643
},
"removed_hours": {
"non_russian_language_tag": 17.215168469899513,
"detected_music_or_singing": 57.27357162351991
},
"recordings": 2319,
"shard_count": 965,
"sample_rates": {
"48000": 278241,
"44100": 36844
},
"language_rule": "language == 'ru'",
"music_rule": "At least two detectors have max targeted AudioSet score >= 0.15; mn40_as resolves disagreements",
"threshold": 0.15,
"primary": {
"model": "EfficientAT/mn10_as",
"weights_sha256": "0bd7dc2443af498c289a2e739f02ebb515d6aa3fd3ab9db539c86123ae368a4e",
"source": {
"repo": "https://github.com/fschmid56/EfficientAT",
"revision": "a425fdce92572e602a1d5634799bd9f1f2efa806"
},
"sample_rate": 32000,
"window_seconds": 10,
"hop_seconds": 5,
"short_window_padding": "zeros",
"target_labels": [
"Music",
"Background music",
"Song",
"Singing",
"Musical instrument"
],
"input_revision": "83c2459778926a7db6144abd552b084de1ac95e9",
"updated_at": "2026-11-12T18:09:36.300543+00:00"
},
"secondary": {
"repo": "MIT/ast-finetuned-audioset-10-10-0.4593",
"revision": "f826b80d28226b62986cc218e5cec390b1096902"
},
"secondary_candidate_rule": "primary_score >= 0.05 OR background_music >= 0.005",
"secondary_checked": 19059,
"negative_audit_samples": 1000,
"negative_audit_majority_music": 0,
"negative_audit_ast_disagreements": 2,
"tie_breaker": {
"model": "EfficientAT/mn40_as",
"weights_sha256": "09624dca1c53128255c65c5a52ff49ae91bb01fe17f0e13b3c3a0c9f24a8478b",
"target_labels": [
"Music",
"Background music",
"Song",
"Singing",
"Musical instrument"
],
"threshold": 0.15,
"total": 1475,
"rule": "Check exactly the samples where primary and AST threshold decisions disagree",
"updated_at": "2026-11-12T23:01:51.415672+00:00"
},
"audio_policy": "Rebuild all shards from local MP3s and audited records; retain native PCM16 audio signal, ASR text and edit spans",
"updated_at": "2026-11-13T00:00:56.107427+00:00",
"construction": "local_mp3_and_audited_records"
},
"transcript_quality": {
"version": 1,
"source_revision": "9c0dc32fc92ef6aa336783e2c838b1c0e3051ae9",
"original_rows": 315085,
"retained_rows": 313655,
"retained_hours": 2055.0831158610463,
"excluded_rows": 1430,
"excluded_hours": 9.52983601662888,
"retained_review_cases": 714,
"independent_asr_checks": 2813,
"letter_rate_reference": {
"reference_rows": 313542,
"median_cps": 10.264134737306962,
"robust_sigma": 0.27782564455327924,
"six_sigma_bounds": [
1.9380879360555046,
54.35896893306615
]
},
"duration_weighted_letters_per_second": 10.388655942318183,
"independent_asr_model": "Qwen/Qwen3-ASR-1.7B",
"asr_revision": "7278e1e70fe206f11671096ffdd38061171dd6e5",
"speech_detector": "Silero v6 (faster-whisper ONNX)",
"policy": "quality/policy.py",
"source_quality_verification_sha256": "0998b09b3a391b31dc91b9980678082abaaab0bf2def7ddd436718c5c235fc5f",
"exclusion_reasons": {
"additional_speech_and_probably_incomplete_transcript": 1204,
"large_independent_asr_disagreement": 226
},
"scope": "Independent checks target extreme rates/severe alignment plus duration-stratified fixed controls; remaining pairs have not had independent ASR verification",
"human_ground_truth_accuracy_not_claimed": true,
"original_text_preserved": true,
"original_retained_audio_preserved": true,
"timestamp_margins_are_not_word_presence_confidence": true
},
"integrated_per_sample_decisions": "filter_scores.parquet"
}
|