Kenpath Labs
Get started

Datasets ready to license.

12 datasets, 680 hours. Play a sample from any card. Every figure was measured from the audio, and the badges say what each dataset is fit for.

9 of 12 datasets · 484 h of speech

  • Scripted call-centre conversations in Hindi, insurance, recorded with each speaker on a separate channel. 356 hours. Transcripts are time-aligned and written in Devanagari script, with English words kept as spoken. Layouts across the set: 1,135 two-channel, 710 one side of a call.

    356 h
    Two channels
    Transcribed
    17% English
    33.7 dB · clean
    • ASR
    • Full duplex
    • Turn-taking
    • Voice agents
    • Diarisation
    • TTS
    490 speakersGet a sample
  • Hindi speech recognition data: 218,428 single-speaker utterance clips of conversational Hindi, call-centre and everyday, each with its own time-aligned transcript in Devanagari script, English words kept as spoken. 350 hours of audio, 3,377,105 transcribed words.

    350 h
    One channel
    Transcribed
    43.5 dB · some background
    • ASR
    • Full duplex
    • Turn-taking
    • Voice agents
    • Diarisation
    • TTS
    667 speakersGet a sample
  • Hindi speaker diarization data: whole call-centre and everyday two-speaker conversations with every speaker turn marked, 432,929 turns in RTTM, for training and scoring who spoke when. 321 hours across 1,394 recordings.

    321 h
    One channel
    Transcribed
    9% English
    30.3 dB · clean
    • ASR
    • Full duplex
    • Turn-taking
    • Voice agents
    • Diarisation
    • TTS
    619 speakersGet a sample
  • Hindi and Tamil full-duplex conversation data: 1,265 two-channel call-centre calls with each speaker on a separate channel, overlaps, backchannels and turn timing preserved, transcripts time-aligned per channel. 264 hours.

    264 h
    Two channels
    Transcribed
    16% English
    29.2 dB · some background
    • ASR
    • Full duplex
    • Turn-taking
    • Voice agents
    • Diarisation
    • TTS
    529 speakersGet a sample
  • Tamil speaker diarization data: whole call-centre and everyday two-speaker conversations with every speaker turn marked, 345,825 turns in RTTM, for training and scoring who spoke when. 229 hours across 1,307 recordings.

    229 h
    One channel
    Transcribed
    39% English
    25.7 dB · some background
    • ASR
    • Full duplex
    • Turn-taking
    • Voice agents
    • Diarisation
    • TTS
    1238 speakersGet a sample
  • Tamil speech recognition data: 157,231 single-speaker utterance clips of conversational Tamil, call-centre and everyday, each with its own time-aligned transcript in Tamil script, English words kept as spoken. 201 hours of audio, 1,748,023 transcribed words.

    201 h
    One channel
    Transcribed
    25.6 dB · some background
    • ASR
    • Full duplex
    • Turn-taking
    • Voice agents
    • Diarisation
    • TTS
    1236 speakersGet a sample
  • Call-centre conversations in Tamil, consumer surveys, recorded on one channel with speakers labelled. 99 hours. Transcripts are time-aligned and written in Tamil script, with English words kept as spoken.

    99 h
    One channel
    Transcribed
    39% English
    22.9 dB · some background
    • ASR
    • Full duplex
    • Turn-taking
    • Voice agents
    • Diarisation
    • TTS
    883 speakersGet a sample
  • Scripted call-centre conversations in Tamil, telecom, delivery, e-commerce and banking, recorded with each speaker on a separate channel. 25 hours. Transcripts are time-aligned and written in Tamil script, with English words kept as spoken. Layouts across the set: 2 one side of a call, 132 two-channel.

    25 h
    Two channels
    Transcribed
    29% English
    31.2 dB · clean
    • ASR
    • Full duplex
    • Turn-taking
    • Voice agents
    • Diarisation
    • TTS
    93 speakersGet a sample
  • Call-centre conversations in Marathi, banking and insurance, recorded on one channel with speakers labelled. 4 hours. Transcripts are time-aligned and written in Devanagari script, with English words kept as spoken.

    4 h
    One channel
    Transcribed
    0% English
    35.5 dB · clean
    • ASR
    • Full duplex
    • Turn-taking
    • Voice agents
    • Diarisation
    • TTS
    18 speakersGet a sample