SL2T is our breakthrough sign language-to-text model powering new features for D
Source
x.com
Author
GoogleDeepMind
Date
Why it matters
Sign-language input is now a shipped phone keyboard modality, and the split design — pose extraction on device, translation on the server — is a reusable pattern for privacy-sensitive multimodalA model that works with more than text — reading images, audio, or video, and sometimes generating them too.Full definition → features.
Terms in this piece · Glossary
multimodal — A model that works with more than text — reading images, audio, or video, and sometimes generating them too.
