Meta’s new AI model can translate speech from more than 100 languages
SMRTR summary
Meta's new AI model, SeamlessM4T, can translate speech between 101 languages with improved accuracy and efficiency. The model uses parallel data mining to associate sounds with matching text in different languages, enabling more direct speech-to-speech translation and potentially paving the way for real-time interpretation in the future.
SMRTR provides this summary for quick context. The original article belongs to MIT Technology Review.
Read the original article