Google DeepMind has introduced a model it names sign-language-to-text, or SL2T, which is designed to translate sign language into written text. According to the company's blog post, the model is now powering new sign language features intended for Deaf and hard of hearing users.
The announcement frames SL2T as a breakthrough and emphasizes putting the capability into users' hands rather than keeping it in the research lab. However, the post excerpt does not include details on supported sign languages, accuracy rates, or how the model was trained.
Because this article is based on a single source, there are no independent reports to compare or contrast. The claims about SL2T currently rest entirely on Google DeepMind's own description.