Vendor page
Introducing speech-to-text, text-to-speech, and more for 1,100+ languages
Meta AI, Meta AI, 2023
Read the originalRead on October 19, 2026
- What it is
- Meta's announcement of the Massively Multilingual Speech (MMS) project on May 22, 2023. The post describes speech-to-text and text-to-speech models covering over 1,100 languages, trained with wav2vec 2.0 and a new labeled dataset.
- How much weight it carries
- Primary announcement from the model's maker, dated on the page. Blog post rather than the peer-reviewed paper, and it reports Meta's own benchmarks, including the comparison with Whisper.
Facts taken from this source
- Meta says its MMS model supports speech-to-text and text-to-speech for 1,107 languages.
“MMS supports speech-to-text and text-to-speech for 1,107 languages and language identification for over 4,000 languages.” (Model description)
- Meta announced the Massively Multilingual Speech (MMS) project on May 22, 2023, with speech-to-text and text-to-speech for over 1,100 languages.
“Introducing speech-to-text, text-to-speech, and more for 1,100+ languages May 22, 2023” (Page headline and date)