DeepMind's SL2T Skips the Gloss Layer, Translating Sign Language to Text on Pixel 11

DeepMind's SL2T model, launched August 12 on the Pixel 11, translates sign language wireframes directly into text. It skips the intermediate "gloss" layer — a rough textual label for each sign — that traditional recognition pipelines rely on. Trained on 100,000+ hours of multilingual data, the approach could lower the cost of adding new language pairs. Only ASL-to-English is available at launch.