Common Voice Spontaneous Speech 2.0 - Kelabit

Kelabit — Kelabit (kzi)

This datasheet is for version 2.0 of the the Mozilla Common Voice Spontaneous Speech dataset for Kelabit (kzi). The dataset contains 1717 clips representing 10 hours of recorded speech (1 hours validated) from 21 speakers.

Language

Kelabit is an Austronesian language in the Apo Duat subgroup. It is spoken by the Kelabit people and is indigenous to Sarawak, Malaysia and North Kalimantan, Indonesia.

Data splits for modelling

Transcriptions

• Prompts: 120

• Duration: 9:50:59 [h:m:s]

• Avg. Transcription Len: 190

• Avg. Duration: 20.65[s]

• Valid Duration: 812.34[s]

• Total hours: 9.85[h]

• Valid hours: 0.23[h]

Writing system

The Kelabit writing system is based on the Latin alphabet.

Symbol table

a e i o u
aa ee ii oo uu
b bp c d dt g gk h j k l m n ng p r s t w y ʼ

Samples

Questions

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

Tulu iko la me’ raut, ngipeh iko pian lem ruma atau ngi luar ruma? Kadi ngudeh?
Kapeh ideh lun merar ngejar/nebada anakadi lem bawang/rang muyuh?
Iye’ nuk mileh lukis lem bawang muyuh nuk iko doo pian?
Tu’en iko mala/mada, nun tibu/bua/pade nuk pengah sah nibu ngan nuk pengah rayeh/laak/sahranih?
Nun a’it ngi bawang iko nuk kenap idih naru tusah ngan muyuh?

Responses

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

Doo' piyah keduih rait ngi luar ih tah narih doo' inan pana narih mirat. Mo ngi luar teh doo'. Kerab teh iko niat nuk doo'-doo' teh beruh.
Lun merar kamih nebada' anak adi' maya' Ao, Bukuh Ago. Kineh ideh nebada' anak adi'.
Mo pasal sistem printah nuuh kai nih  tuseh keriit lah 'am tidih doo' tuuh kadi' lat ngilad dih naem doo' mimang ideh la' naru kai nuk tusah nih ngiyuk kai tusah lah. Kayu' pasal tana nih. Mileh-mileh nideh naru' dih baken. 
Tibu nuk [err] narih nibu paad-paad ideh inan bua' kayu' ineh bua' kiran, bua' datu', bua' lengaat mey bua' langsat kayu' ineh. Neh tibu nuk [err] pengeh narih neh nibu dih. 
Mula' naan teh ait ngen tauh kinih. Luk pelaba biasa ngi ruma tauh mait uluh nideh mait keted, mait aleb ngen lun merar dih. Anak adi' iih pelaba mai batek nidah. Mula' naan ait tauh kinih. Kadi'  nah tauh perlu may Ruma' Sakit nabat. 

Recommended post-processing

The questions use the ’ character, while the responses use the ' character instead. These characters should be treated as equivalent.

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

Get involved!

Community links

• Common Voice translators on Pontoon

Contribute

• Contribute questions

• Validate questions

• Answer questions

• Transcribe recordings

• Validate transcriptions

Acknowledgements

Funding

This dataset was partially funded by the Open Multilingual Speech Fund managed by Mozilla Common Voice.

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 demographics spec. These will only be reported if the speaker opted in to provide that information. ↩ ↩2