Putting sign language AI into users’ hands
Summary
Google DeepMind introduces sign-language-to-text SL2T, a breakthrough model powering sign-to-text features in Gboard and Live Transcribe, starting with ASL to English. The model is trained on over 100,000 hours across 50+ sign languages and uses on-device pose landmark data to protect privacy, translating directly from landmarks to text. The project emphasizes community governance with AISLAC and outlines plans for broader language support and capabilities.