Google Puts Sign Language to Text AI in Pixel 11
The Pixel 11 launch brings sign language translation out of research and into a consumer device, giving deaf users a typing alternative in everyday apps.
Reporting from 1 source: GIGAZINE.
Google DeepMind announced a multilingual sign-language-to-text translation model called SL2T that converts sign language video into text. The model will ship in the Pixel 11, releasing August 20, 2026, and will let users input text through sign language in Gboard and Live Transcribe, starting with American Sign Language to English at no extra charge.
Google DeepMind trained SL2T on more than 100,000 hours of video covering over 50 sign languages. The model reads body movements, converts them into coordinates, and translates them, a process the team says was difficult because each sign language has its own grammar and vocabulary and accurate tracking at high frame rates is hard.
The model also addresses practical issues: minimizing latency during streaming, preventing hallucinations when input is not sign language, and improving performance for one-handed signing. Test participants reported that signing is faster, more natural, and more comfortable than typing text.
SL2T first appears in the Pixel 11's Gboard text input app and the Live Transcribe app. Initial support covers American Sign Language to English only, with more devices to follow. The feature costs nothing extra.
Synthesized by Yomimono from the 1 cited source below, including Japanese-language reporting where cited, then editorially reviewed before publishing.
Sources
- GIGAZINE Googleが手話をテキストに変換するAIモデルをPixel 11に搭載