Google Pixel 11 Series Launches Sign Language-to-Text Feature

On August 13, DeepMind, Google's artificial intelligence lab, announced the launch of a large-scale multilingual Sign Language to Text (SL2T) translation model. The feature is now available in the Gboard and Live Transcribe apps on Pixel 11 series smartphones. Initially supporting American Sign Language (ASL) only, the system will expand to additional sign languages and devices in the future. Trained on over 100,000 hours of data covering more than 50 sign languages, the model processes sign language by converting it into positional data of pose keypoints—without uploading raw video streams—to protect user privacy. The system directly interprets spatial information, bypassing the commonly used "intermediate annotation" approach. Globally, approximately 70 million deaf and hard-of-hearing individuals use more than 200 distinct sign languages. Testing has shown that inputting via ASL is faster, more natural, and more enjoyable than typing in English. DeepMind emphasized that sign languages are full-fledged natural languages with their own grammar and vocabulary, requiring genuine machine translation techniques for accurate interpretation.