Datasets:
Tasks:
Automatic Speech Recognition
Modalities:
Audio
Formats:
soundfolder
Languages:
Kamba (Kenya)
Size:
< 1K
License:
Search is not available for this dataset
audio audioduration (s) 4.27 8.97 |
|---|
Kikamba First Open Corpus - First for 5M speakers in Kenya
14 WAV files, 16kHz, CC-BY-4.0 Creator: Kamba Translation Team, est. 22/6/2020 Kitui - 12 core, 100+ network (Kitui, Machakos, Makueni, Nairobi, diaspora) Dialects: Masaku, Kitui, Makueni Consent: Ethically collected with verbal consent in Kikamba Goal: Grow to 500 voices for Mozilla Common Voice
Submitted to Mozilla Data Collective 30 Sept 2026. Seeking NLP collaborators & funding.
- Downloads last month
- 84