Mozilla has announced the release of the 18th dataset of Common Voice, which is now available for download. This dataset is part of Mozilla's strategy to ensure broader access to voice technologies. It is a free dataset of multilingual voice fragments and associated text data, released under a CC0 (public domain) license. The creation of this dataset is a collaborative effort of the community, including participants providing voice and text, language activists, technologists, researchers, and other members of the Common Voice community.
The total volume of the Common Voice dataset has reached 31,841 hours, of which 20,789 hours of speech data have been verified by the community. This is an increase of 700 hours of speech data compared to the last dataset release, and 381 hours more of newly verified data. The 18th dataset consists of fragments in 129 languages, including 5 new languages added in this release.
The new dataset features five new languages: Xhosa (South Africa), Kalenjin (Kenya), Kidaw’ida (Kenya), Dholuo (Kenya and Tanzania), and Setswana (Botswana, Zimbabwe, Namibia, South Africa). Hundreds of millions of people worldwide speak these languages and can now receive better support in voice technologies.
If you are interested in Common Voice, there are many ways to join the community of contributors. You can share your voice or write and submit original suggestions in your language to help create the next dataset. If your language is not yet in Common Voice, you can request its addition using a special form. Technical contributions to the open project on GitHub are also welcome.
Mozilla is always eager to receive feedback on new releases. You can contact them in the Common Voice forums, chat with them on Matrix, or reach out to the team directly at commonvoice@mozilla.com. They are particularly interested in learning about what users of the datasets create or explore using the dataset. A better understanding of the dataset users' needs can help them determine the direction that better meets user requirements.
Source: linux.org.ru
