Common Voice Spontaneous Speech 1.0 - Kelabit

Locale: kzi

Size: 193.90 MB

Task: ASR

Format: MP3

License: CC-0

Kelabit — Kelabit (kzi)

This datasheet has been generated automatically, we would love to include more information, if you would like to help out, get in touch!

This datasheet is for version 23.0 of the the Mozilla Common Voice Spontaneous Speech dataset for Kelabit (kzi). The dataset contains 1741 clips representing 10 hours of recorded speech (1 hours validated) from 21 speakers.

Demographic information

The dataset includes the following distribution of age and gender.

Gender

Self-declared gender information, frequency refers to the number of clips annotated with this gender.

Age

Self-declared age information, frequency refers to the number of clips annotated with this age band.

Data splits for modelling

SplitCount
Train1112
Test296
Dev333

Transcriptions

  • Prompts: 120

  • Duration: 34662672[ms]

  • Avg. Transcription Len: 190

  • Avg. Duration: 19.91[s]

  • Valid Duration: 812.34[s]

  • Total hours: 9.63[h]

  • Valid hours: 0.23[h]

Questions

There follows a randomly selected sample of questions used in the corpus.

Nun naru lun muyuh bubuh me’ ngi bawang beken? / Kadi ngudah lun muyuh bubuh me’ ngi bawang beken?
Enun program radio lem bawang nuk iko doo pian?
Ken inan bajet senaru/senulis muyuh dengeruyung? / Ken naru muyuh dengeruyung bajet?
Iko seritah lawe iko nier wayang ngi ruma wayang lem bawang muyuh?
Kapeh ayu edto ngi nuk inan iko mudeng?
Responses

There follows a randomly selected sample of transcribed responses from the corpus.

Mo raut ebol, raut ebol, raut volley ball, raut gaing nideh ngilad, raut pepana' nideh, raut guli nideh ngilad.
Manid na'an teh pakaian doo' piyan tauh lemulun eh. Maya' lawey mudeng ngi ruma' beto' narih keh pakai nuk buyur neh narih. La' mey lati' beto' narih keh pakai nuk senang koh mey lati' nikoh. La' mey kerja ngi upis beto' koh keh la' pakai gaun neh narih. Maya' kerja nuk tu'en narih teh sapa' nuk doo' piyan pakai narih.
Teknologi nuk inan uih rapang  i h ineh inan ibal lemulun ngidih nuk merey  karuh, merey talk leh. Bisnes. Kinih-kinih kedeh , kinih-kinih kedeh nuru'  arih masuk bisnes bah. Err... ta'ut narih .  Mula' lem...lem  pengeh uih ni'er ibal lemulun deh masuk bisnes neh keyh. Mileh' usin deh raruh bah. 'di keduih rapang ngelinuh lemulun nuk nasihat tauh mala kinih-kinih-kinih tauh. Beng, beng, raut beng kinih usin pakai tauh kedeh ko'. Masuk nideh beng, masuk nideh na'...inan na' nideh pa... inan peruyung lah gen bisnes-bisnes. Mileh raruh teh usin dedih bah. Kadi' keduih  taut lah, kadi' keduih rapang masuk nuk tad-tad ineh neh bah. 'Ey' neto keduih nekinih masuk naru' sineh neto'.
Na'em ngadan getu'en keli' keduih. Na'em men lun merar mala ngadan getu'en malam, mala idih getu'en tupu tideh nuk mudeng liyang langit nangey  senaru' Tuhan Alla, neh tupu nidih, erang gawa' ngih. Mo.
[noise] Moq kanap so karu'uih naruh bajet, [a..]lam [dat..] daih kamih[a..] lam kamih ngaruyung, masti narih inan bajet dih tah do, naam narih pakai usin mu'oh mu'oh, dih tah narih mi'lah gi'rah usin, ripun idih man...

Fields

Each row of a tsv file represents a single audio clip, and contains the following information:

  • client_id - hashed UUID of a given user

  • audio_id - numeric id for audio file

  • audio_file - audio file name

  • duration_ms - duration of audio in milliseconds

  • prompt_id - numeric id for prompt

  • prompt - question for user

  • transcription - transcription of the audio response

  • votes - number of people that who approved a given transcript

  • age - age of the speaker1

  • gender - gender of the speaker1

  • language - language name

  • split - for data modelling, which subset of the data does this clip pertain to

  • char_per_sec - how many characters of transcription per second of audio

  • quality_tags - some automated assessment of the transcription--audio pair, separated by |

    • transcription-length - character per second under 3 characters per second

    • speech-rate - characters per second over 30 characters per second

    • short-audio - audio length under 2 seconds

    • long-audio - audio length over 30 seconds

Community links

Contribute

Licence

This dataset is released under the Creative Commons Zero (CC-0) licence. By downloading this data you agree to not determine the identity of speakers in the dataset.

Footnotes

  1. For a full list of age, gender, and accent options, see the demograpics spec. These will only be reported if the speaker opted in to provide that information. ↩ ↩2