{ "id": "LK-SP-TAM-003", "slug": "tamil-general-conversation", "title": "Tamil general conversation", "modality": "speech", "speechType": "conversational", "kind": "conversations", "unit": "conversations", "collection": "general-conversation", "language": { "name": "Tamil", "native": "தமிழ்", "iso": "tam", "bcp47": "ta-IN", "script": "Tamil" }, "languages": [ { "name": "Tamil", "native": "தமிழ்", "iso": "tam", "bcp47": "ta-IN", "script": "Tamil" } ], "region": "India", "domain": "Open topic", "scripted": false, "parties": null, "hours": 105.4, "transcribed": true, "channels": "mono", "source": "Recorded for the dataset", "review": "Every recording has a complete, segment-level transcript", "personalData": "Redacted", "licence": "Custom", "sheetRows": [ 4 ], "sample": { "key": "conversations-ta-gc-99", "slug": "tamil-general-conversation", "dual": false, "audio": { "sampleRateHz": 16000, "bitDepth": 16, "codec": "pcm_s16le", "container": "flac", "filesPerConversation": 1, "measured": true }, "metrics": { "durationS": 1478.7, "conversations": 6, "turns": 384, "words": 3333, "talkTimeS": { "1": 825, "2": 566.7 }, "speechShare": 0.941, "overlapShare": 0, "silenceShare": 0.059, "wordsPerMinute": 143.7, "turnDurationS": { "median": 2.58, "p90": 7.98 }, "floorTransfer": { "count": 202, "medianS": 0, "p10S": 0, "p90S": 0.929, "overlappedShare": 0, "histogram": { "edgesS": [ -2, -1, -0.5, -0.2, 0, 0.2, 0.5, 1, 2, 4 ], "counts": [ 0, 0, 0, 0, 0, 172, 0, 12, 11, 4, 3 ] } }, "backchannels": { "count": 61, "perMinute": 2.48 }, "codeMix": { "latinTokens": 1188, "nativeTokens": 2143, "latinShare": 0.357 }, "events": { "[filler]": 108, "[silence]": 3, "[unintelligible]": 43, "[suppressed]": 4 }, "overlapEvents": 0 }, "quality": { "snrDb": { "median": 42.1, "min": 29.4, "max": 56.4 }, "noiseFloorDb": -60.2, "speechLevelDb": -15.6, "bandwidthHz": 7950, "clippingShare": 0, "label": "clean", "dnsmos": { "sig": 3, "bak": 3, "ovrl": 2.33 }, "method": "Frame RMS at 20 ms on the channels mixed to mono. Noise floor is the 10th percentile, speech level the 90th; SNR is their difference. Bandwidth is the highest frequency at which speech still rises 6 dB above the recording's own noise spectrum." }, "speakers": [ { "role": "Speaker 1", "id": "ta-gc-99-s0041", "gender": "female", "ageBand": null }, { "role": "Speaker 2", "id": "ta-gc-99-s0113", "gender": "female", "ageBand": null } ], "speakerSummary": { "distinctIds": 12, "genders": { "female": 9, "male": 3 }, "ageBands": null }, "conversations": [ { "index": 1, "sourceId": "conv-a3c409b986", "title": "Open conversation", "durationS": 246.7, "turns": 40, "words": 580, "speakers": [ { "role": "Speaker 1", "id": "ta-gc-99-s0041", "gender": "female", "ageBand": null }, { "role": "Speaker 2", "id": "ta-gc-99-s0113", "gender": "female", "ageBand": null } ], "flagged": 0, "silenceShare": 0, "overlapShare": 0, "latinShare": 0.357, "wordsPerMinute": 141.1, "specimen": null, "quality": { "noiseFloorDb": -59.6, "speechLevelDb": -11.7, "snrDb": 47.9, "clippingShare": 0, "bandwidthHz": 8000 }, "qualityLabel": "clean", "excerptSnrDb": 51.7 }, { "index": 2, "sourceId": "conv-fb7c7e4e4f", "title": "Open conversation", "durationS": 239.2, "turns": 66, "words": 549, "speakers": [ { "role": "Speaker 1", "id": "ta-gc-99-s0064", "gender": "female", "ageBand": null }, { "role": "Speaker 2", "id": "ta-gc-99-s0151", "gender": "male", "ageBand": null } ], "flagged": 0, "silenceShare": 0, "overlapShare": 0, "latinShare": 0.477, "wordsPerMinute": 137.7, "specimen": { "id": "tamil-general-conversation--4", "windowS": 43.453, "offsetS": 61.411, "segments": 14, "bytes": 282653, "sha256": "e5156a130e789b7480f4b9cf999b858fec88069501d9495e82730dbd71743be1", "dnsmos": { "sig": 3.4, "bak": 2.46, "ovrl": 2.41 } }, "quality": { "noiseFloorDb": -44.2, "speechLevelDb": -14.8, "snrDb": 29.4, "clippingShare": 0, "bandwidthHz": 8000 }, "qualityLabel": "some background", "excerptSnrDb": 28 }, { "index": 3, "sourceId": "conv-ba19422555", "title": "Open conversation", "durationS": 248.1, "turns": 83, "words": 563, "speakers": [ { "role": "Speaker 1", "id": "ta-gc-99-s0176", "gender": "female", "ageBand": null }, { "role": "Speaker 2", "id": "ta-gc-99-s0177", "gender": "female", "ageBand": null } ], "flagged": 0, "silenceShare": 0.121, "overlapShare": 0, "latinShare": 0.361, "wordsPerMinute": 154.8, "specimen": { "id": "tamil-general-conversation--1", "windowS": 44.686, "offsetS": 0, "segments": 19, "bytes": 279885, "sha256": "5f6ca60c1d7ed0e2558383d4cbe91fab14a5be6c3f80729d49c058744a8fbba7", "dnsmos": { "sig": 2.63, "bak": 3.22, "ovrl": 2.17 } }, "quality": { "noiseFloorDb": -60.8, "speechLevelDb": -24.8, "snrDb": 36.1, "clippingShare": 0, "bandwidthHz": 7900 }, "qualityLabel": "clean", "excerptSnrDb": 37.1 }, { "index": 4, "sourceId": "conv-e1da4b6c7a", "title": "Open conversation", "durationS": 248.4, "turns": 82, "words": 502, "speakers": [ { "role": "Speaker 1", "id": "ta-gc-99-s0122", "gender": "male", "ageBand": null }, { "role": "Speaker 2", "id": "ta-gc-99-s0087", "gender": "female", "ageBand": null } ], "flagged": 0, "silenceShare": 0.158, "overlapShare": 0, "latinShare": 0.371, "wordsPerMinute": 144.1, "specimen": { "id": "tamil-general-conversation--3", "windowS": 43.973, "offsetS": 30.119, "segments": 18, "bytes": 277216, "sha256": "c9acd1c66a11e3b0d01d493343f44ae7f945e104861db205f9738aed1b42a0a9", "dnsmos": { "sig": 3.11, "bak": 2.76, "ovrl": 2.27 } }, "quality": { "noiseFloorDb": -57.2, "speechLevelDb": -16.4, "snrDb": 40.8, "clippingShare": 0, "bandwidthHz": 7900 }, "qualityLabel": "clean", "excerptSnrDb": 29.2 }, { "index": 5, "sourceId": "conv-04ba9496c8", "title": "Open conversation", "durationS": 248.4, "turns": 64, "words": 659, "speakers": [ { "role": "Speaker 1", "id": "ta-gc-99-s0057", "gender": "male", "ageBand": null }, { "role": "Speaker 2", "id": "ta-gc-99-s0211", "gender": "female", "ageBand": null } ], "flagged": 0, "silenceShare": 0.054, "overlapShare": 0, "latinShare": 0.302, "wordsPerMinute": 168.3, "specimen": { "id": "tamil-general-conversation--2", "windowS": 41.389, "offsetS": 0, "segments": 18, "bytes": 265996, "sha256": "a4e0b13600118823602c2c1e2473d88e322f194bcd5ab58e6f061ba6263cdf78", "dnsmos": { "sig": 2.85, "bak": 3.54, "ovrl": 2.48 } }, "quality": { "noiseFloorDb": -68.9, "speechLevelDb": -12.5, "snrDb": 56.4, "clippingShare": 0, "bandwidthHz": 8000 }, "qualityLabel": "clean", "excerptSnrDb": 57.9 }, { "index": 6, "sourceId": "conv-7a7bb2a2de", "title": "Open conversation", "durationS": 247.9, "turns": 49, "words": 480, "speakers": [ { "role": "Speaker 1", "id": "ta-gc-99-s0248", "gender": "female", "ageBand": null }, { "role": "Speaker 2", "id": "ta-gc-99-s0247", "gender": "female", "ageBand": null } ], "flagged": 1, "silenceShare": 0.017, "overlapShare": 0, "latinShare": 0.273, "wordsPerMinute": 118.2, "specimen": null, "quality": { "noiseFloorDb": -66.5, "speechLevelDb": -23, "snrDb": 43.5, "clippingShare": 0, "bandwidthHz": 7900 }, "qualityLabel": "clean", "excerptSnrDb": 44.6 } ], "specimens": [ "tamil-general-conversation--1", "tamil-general-conversation--2", "tamil-general-conversation--3", "tamil-general-conversation--4" ], "transcriptFormat": "json", "piiFlaggedSegments": 1, "withheld": false }, "utterances": null, "gaps": [], "bundleId": "conversations-ta-gc-99", "relatedSlugs": [ "tamil-speech-recognition-utterances", "tamil-speaker-diarization" ], "measured": { "id": "ta-gc-99", "files": 464, "conversations": 464, "channels": "464 mono conversation", "bandwidth": "wideband (8 kHz)", "bandwidthCounts": { "wideband (8 kHz)": 464 }, "telephoneBandFiles": 0, "utterances": null, "turns": null, "speakers": 279, "speakerGender": { "female": 208, "male": 47 }, "transcripts": 464, "segments": 79560, "words": 958275, "snrMedianDb": 25.7, "container": "flac", "sampleRateHz": 16000, "bitDepth": 16, "pii": "Numbers of 4 or more digits, digit strings spoken as words, and email addresses are masked as [pii] in text and replaced by a tone in audio. First names are not masked.", "release": "v1.0", "updatedAt": "2026-09-28T23:34", "profile": { "conversations": 464, "medianConversationMin": 14.67, "talkShareByRank": [ 0.673, 0.327 ], "silenceShare": 0.173, "overlapShare": 0, "turnsPerMin": 8.96, "wordsPerMin": 151.5, "codeMixShare": 0.328, "twoChannel": null }, "provisional": false }, "url": "https://kenpathlabs.com/lokah/datasets/tamil-general-conversation" }