{ "id": "LK-SP-MAR-001", "slug": "marathi-general-conversation", "title": "Marathi general conversation", "modality": "speech", "speechType": "conversational", "kind": "conversations", "unit": "conversations", "collection": "general-conversation", "language": { "name": "Marathi", "native": "मराठी", "iso": "mar", "bcp47": "mr-IN", "script": "Devanagari" }, "languages": [ { "name": "Marathi", "native": "मराठी", "iso": "mar", "bcp47": "mr-IN", "script": "Devanagari" } ], "region": "India", "domain": "Open topic", "scripted": false, "parties": null, "hours": 9.4, "transcribed": true, "channels": "mono", "source": "Recorded for the dataset", "review": "Every recording has a complete, segment-level transcript", "personalData": "Redacted", "licence": "Custom", "sheetRows": [ 15 ], "sample": { "key": "conversations-mr-gc", "slug": "marathi-general-conversation", "dual": false, "audio": { "sampleRateHz": 16000, "bitDepth": 16, "codec": "pcm_s16le", "container": "flac", "filesPerConversation": 1, "measured": true }, "metrics": { "durationS": 1496.6, "conversations": 6, "turns": 428, "words": 2877, "talkTimeS": { "1": 621.2, "2": 648.6 }, "speechShare": 0.834, "overlapShare": 0.015, "silenceShare": 0.166, "wordsPerMinute": 138.3, "turnDurationS": { "median": 2.5, "p90": 5.94 }, "floorTransfer": { "count": 149, "medianS": 0.803, "p10S": 0.156, "p90S": 1.573, "overlappedShare": 0.013, "histogram": { "edgesS": [ -2, -1, -0.5, -0.2, 0, 0.2, 0.5, 1, 2, 4 ], "counts": [ 0, 0, 1, 0, 1, 18, 30, 47, 46, 4, 2 ] } }, "backchannels": { "count": 97, "perMinute": 3.89 }, "codeMix": { "latinTokens": 0, "nativeTokens": 2877, "latinShare": 0 }, "events": { "[unclear]": 62, "[coughing]": 7, "[laughing]": 2 }, "overlapEvents": 0 }, "quality": { "snrDb": { "median": 60, "min": 48.5, "max": 102.1 }, "noiseFloorDb": -74.9, "speechLevelDb": -15.7, "bandwidthHz": 8000, "clippingShare": 0.03004, "label": "some background", "dnsmos": { "sig": 3, "bak": 3.48, "ovrl": 2.55 }, "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": "mr-gc-s0009", "gender": "male", "ageBand": null }, { "role": "Speaker 2", "id": "mr-gc-s0010", "gender": "male", "ageBand": null } ], "speakerSummary": { "distinctIds": 12, "genders": { "male": 6, "female": 5 }, "ageBands": null }, "conversations": [ { "index": 1, "sourceId": "conv-4a9d0e3d07", "title": "Open conversation", "durationS": 249.6, "turns": 82, "words": 507, "speakers": [ { "role": "Speaker 1", "id": "mr-gc-s0009", "gender": "male", "ageBand": null }, { "role": "Speaker 2", "id": "mr-gc-s0010", "gender": "male", "ageBand": null } ], "flagged": 0, "silenceShare": 0.136, "overlapShare": 0.016, "latinShare": 0, "wordsPerMinute": 141, "specimen": { "id": "marathi-general-conversation--1", "windowS": 44.767, "offsetS": 191.325, "segments": 20, "bytes": 276821, "sha256": "afdd448433a6ee118b59e364eea170bac31298b30a9bad1d0c3bd4234233361d", "dnsmos": { "sig": 3.19, "bak": 3.6, "ovrl": 2.73 } }, "quality": { "noiseFloorDb": -68.3, "speechLevelDb": -16.8, "snrDb": 51.5, "clippingShare": 0, "bandwidthHz": 5600 }, "qualityLabel": "clean", "excerptSnrDb": 47.4 }, { "index": 2, "sourceId": "conv-7f955a3606", "title": "Open conversation", "durationS": 247.8, "turns": 78, "words": 505, "speakers": [ { "role": "Speaker 1", "id": "mr-gc-s0019", "gender": "male", "ageBand": null }, { "role": "Speaker 2", "id": "mr-gc-s0020", "gender": "female", "ageBand": null } ], "flagged": 0, "silenceShare": 0.168, "overlapShare": 0.02, "latinShare": 0, "wordsPerMinute": 147, "specimen": { "id": "marathi-general-conversation--3", "windowS": 43.542, "offsetS": 3.02, "segments": 17, "bytes": 273508, "sha256": "b5e197fc1f8e4ad3d2f6a7ba992fae137b961235e4d345f02759248f311492d8", "dnsmos": { "sig": 2.74, "bak": 3.62, "ovrl": 2.41 } }, "quality": { "noiseFloorDb": -87, "speechLevelDb": -13.8, "snrDb": 73.2, "clippingShare": 0, "bandwidthHz": 8000 }, "qualityLabel": "clean", "excerptSnrDb": 90.5 }, { "index": 3, "sourceId": "conv-d8db318cd9", "title": "Open conversation", "durationS": 249.7, "turns": 71, "words": 511, "speakers": [ { "role": "Speaker 1", "id": "mr-gc-s0007", "gender": "female", "ageBand": null }, { "role": "Speaker 2", "id": "mr-gc-s0026", "gender": null, "ageBand": null } ], "flagged": 0, "silenceShare": 0.106, "overlapShare": 0.036, "latinShare": 0, "wordsPerMinute": 137.4, "specimen": { "id": "marathi-general-conversation--4", "windowS": 42.623, "offsetS": 0.217, "segments": 16, "bytes": 262468, "sha256": "b842a7fe44bf513b65b90e73631f2c2308269fb8c3a83b178b7b96d2241b5ee6", "dnsmos": { "sig": 2.96, "bak": 3.51, "ovrl": 2.53 } }, "quality": { "noiseFloorDb": -81.5, "speechLevelDb": -14.6, "snrDb": 66.8, "clippingShare": 0, "bandwidthHz": 8000 }, "qualityLabel": "clean", "excerptSnrDb": 66.8 }, { "index": 4, "sourceId": "conv-0f8d9ae097", "title": "Open conversation", "durationS": 249.6, "turns": 54, "words": 467, "speakers": [ { "role": "Speaker 1", "id": "mr-gc-s0027", "gender": "female", "ageBand": null }, { "role": "Speaker 2", "id": "mr-gc-s0028", "gender": "female", "ageBand": null } ], "flagged": 0, "silenceShare": 0.11, "overlapShare": 0, "latinShare": 0, "wordsPerMinute": 126.2, "specimen": null, "quality": { "noiseFloorDb": -56.1, "speechLevelDb": -3, "snrDb": 53.1, "clippingShare": 0.03004, "bandwidthHz": 8000 }, "qualityLabel": "some background", "excerptSnrDb": 55.6 }, { "index": 5, "sourceId": "conv-fb45dd233e", "title": "Open conversation", "durationS": 249.9, "turns": 83, "words": 460, "speakers": [ { "role": "Speaker 1", "id": "mr-gc-s0030", "gender": "female", "ageBand": null }, { "role": "Speaker 2", "id": "mr-gc-s0029", "gender": "male", "ageBand": null } ], "flagged": 0, "silenceShare": 0.295, "overlapShare": 0.013, "latinShare": 0, "wordsPerMinute": 156.5, "specimen": { "id": "marathi-general-conversation--2", "windowS": 44.024, "offsetS": 85.551, "segments": 18, "bytes": 270421, "sha256": "4dab4cbffd4e034779f0a76fa99072961f0ccaba0fb4153a6707303ce53acf05", "dnsmos": { "sig": 3.09, "bak": 3.2, "ovrl": 2.53 } }, "quality": { "noiseFloorDb": -68.1, "speechLevelDb": -19.6, "snrDb": 48.5, "clippingShare": 0, "bandwidthHz": 7600 }, "qualityLabel": "clean", "excerptSnrDb": 50.4 }, { "index": 6, "sourceId": "conv-c96ba9f4f1", "title": "Open conversation", "durationS": 250, "turns": 60, "words": 427, "speakers": [ { "role": "Speaker 1", "id": "mr-gc-s0003", "gender": "male", "ageBand": null }, { "role": "Speaker 2", "id": "mr-gc-s0004", "gender": "male", "ageBand": null } ], "flagged": 0, "silenceShare": 0.182, "overlapShare": 0.002, "latinShare": 0, "wordsPerMinute": 125.2, "specimen": null, "quality": { "noiseFloorDb": -120, "speechLevelDb": -17.9, "snrDb": 102.1, "clippingShare": 0, "bandwidthHz": 8000 }, "qualityLabel": "clean", "excerptSnrDb": 75.1 } ], "specimens": [ "marathi-general-conversation--1", "marathi-general-conversation--2", "marathi-general-conversation--3", "marathi-general-conversation--4" ], "transcriptFormat": "json", "piiFlaggedSegments": 0, "withheld": false }, "utterances": null, "gaps": [], "bundleId": "conversations-mr-gc", "relatedSlugs": [], "measured": { "id": "mr-gc", "files": 36, "conversations": 36, "channels": "36 mono conversation", "bandwidth": "telephone (4 kHz)", "bandwidthCounts": { "telephone (4 kHz)": 24, "wideband (8 kHz)": 12 }, "telephoneBandFiles": 9, "utterances": null, "turns": null, "speakers": 40, "speakerGender": { "female": 21, "male": 15 }, "transcripts": 36, "segments": 10932, "words": 74091, "snrMedianDb": 37.8, "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": "2026-09-29T02:19", "profile": { "conversations": 36, "medianConversationMin": 15.56, "talkShareByRank": [ 0.629, 0.371 ], "silenceShare": 0.167, "overlapShare": 0.044, "turnsPerMin": 10.61, "wordsPerMin": 129, "codeMixShare": 0, "twoChannel": null }, "provisional": false }, "url": "https://kenpathlabs.com/lokah/datasets/marathi-general-conversation" }