Contribute your language
AI tools often fall short in languages with little available data. If you hold text or speech data in your language, here's how to share it.
Ways to contribute
Share a dataset as an uploader
For a dataset you have the rights to. Request uploader access, we review it, then you publish under your own name or organisation.
Become an uploaderRecord or validate on Common Voice
No dataset yet? Donate your voice or check other people's recordings on Mozilla Common Voice.
Go to Common VoiceTalk to our team
For archives, institutions and language communities who want to discuss licensing, preparation or a partnership first.
Get in touchWhat helps most
Speech recordings, ideally with transcripts
Text in the language: books, news, forums, community writing
Translations paired with another language
Dictionaries, word lists and pronunciation guides
Before you share
Make sure you have the right to share it, and the consent of the people in it.
Choose how it can be used: openly licensed, community-governed or compensated.
Prepare a small sample so others can judge it before downloading.
Not sure if your language is covered yet? Search the datasets before you start.