Google DeepMind Unveils SL2T Sign Language Model, Coming to Pixel 11

heiga_zen · x · 2026-08-13

Google DeepMind and the Android team introduced the sign language translation model "SL2T," officially integrated into Pixel 11's Gboard and Live Transcribe. The feature allows users to input text directly into their devices using sign language.

Sign language involves the coordination of facial expressions and upper body movements, not just gestures. SL2T achieves a breakthrough by accurately capturing these elements and translating them directly into text. Developed in deep collaboration with Deaf engineers and communities, the project features a privacy-preserving design that processes only skeletal landmarks without retaining camera footage, and supports sign language input even while holding the phone with one hand.

Related event: DeepMind Unveils SL2T Sign Language-to-Text Model, Debuts on Pixel 11(14 posts)→

Original post →

More from Multimodal

Multimodal channel →