Fact · Languages and accents
1.3BLEU points
The SeamlessM4T paper says that, compared with strong cascaded models, its speech to text translation into English is 1.3 BLEU points better.
Checked against its source on October 17, 2026
In the source's own words
“Compared to strong cascaded models, SeamlessM4T improves the quality of into-English translation by 1.3 BLEU points in speech-to-text”
SeamlessM4T: Massively Multilingual and Multimodal Machine Translation, Seamless Communication, Meta AI (arXiv), 2023, Abstract.
Open the original (DOI 10.48550/arXiv.2308.11596)- What it measures
- A cascade here means a speech recognizer followed by a text translation model. SeamlessM4T is one model that translates speech directly.
- What it does not show
- The gain is measured on the paper's own benchmark. It is a few BLEU points, which is small next to the errors a cascade can carry.