SL2T is a multilingual sign-language-to-text translation model now powering sign-to-text dictation in Gboard and Live Transcribe on Pixel 11, starting with ASL to English. Trained on over 100,000 hours across more than 50 sign languages, SL2T translates pose landmark coordinates directly into text, bypassing intermediate glosses. The model achieves 70 BLEURT on the FLEURS-ASL benchmark and lets users sign to search, draft messages, or ask Gemini queries.
No score is assigned. Sources and their independence are shown in the citation chain below.