Bullish

Google Integrates Sign Language Input Model Into Pixel 11 Gboard

18:52

DeepMind's SL2T model enables ASL-to-text input on Pixel 11 via Gboard, processing over 100,000 hours of data with privacy-preserving coordinate transmission.

Woofun AI reports that Google DeepMind has integrated the SL2T sign language-to-text model into Gboard and Live Transcribe on the Pixel 11. The system currently translates American Sign Language into English, allowing users to search, type messages, and query Gemini using hand gestures. Trained on over 100,000 hours of data covering more than 50 sign language types, the model processes only extracted coordinate data for hands, face, and body movements, ensuring original video is not uploaded to servers.

WOOFUN AI

Impact Assessment · Quick Read

This integration marks a significant step in accessibility technology, potentially expanding the addressable market for mobile input methods. By leveraging on-device coordinate extraction rather than raw video, Google addresses privacy concerns while enabling real-time translation. The focus on ASL initially may drive adoption among deaf communities, though support for other sign languages could follow given the diverse training dataset.
Generated by WOOFUN AI · For reference only, not investment advice

Comments

Me
Replying to @User
0/800

No comments yet.

Notifications

Sign in to view messages
View all messagesManage subscriptions