Release Notes
Common Voice is thrilled to announce the release of its 18th dataset, now available for download. Committed to making voice technologies more accessible, this release offers a cost and copyright-free dataset of multilingual voice clips and associated text data under a CC0 license. The dataset, driven by community contributors, includes 31,841 hours of speech data, with 20,789 hours validated by the community. This marks an increase of nearly 700 hours since the last release, featuring clips from 128 languages, including new additions Xhosa, Kalenjin, Kidaw'ida, Dholuo, and Setswana.
Check out the official blog post!
What's Changed
- feat: add cv corpus v18 by @moz-dfeller in #4517
Full Changelog: release-v1.121.0...release-v1.121.1