DeepMind has announced the release of sign-language-to-text (SL2T), a new artificial intelligence model designed to improve accessibility for Deaf and hard of hearing users. According to DeepMind, the model represents a breakthrough in assistive technology and will power new sign language features.
The SL2T model is capable of translating sign language into written text, potentially enabling better communication between sign language users and those who don’t understand signing. DeepMind describes this as putting “sign language AI into users’ hands,” suggesting the technology will be made available for practical use.
While DeepMind characterizes the model as a breakthrough, specific details about the technology’s accuracy, supported sign languages, or deployment timeline were not provided in the announcement. The development represents DeepMind’s continued focus on applying artificial intelligence to real-world accessibility challenges for underserved communities.