Google’s new Pixel 11 series devices bring more than just hardware upgrades. One of the lineup’s most interesting new features is designed to break down the communication barrier between deaf and hard-of-hearing users and those who don’t know sign language.
Pixel Camera Can Now Turn Sign Language Into Text
Google has introduced a groundbreaking accessibility feature that uses the Pixel Camera to recognize sign language and automatically convert it into text. This innovative system allows users to sign as they normally would without the interruption of typing responses. With this new capability, Google aims to transform smartphones from passive listening devices into active participants in natural, two-way conversations during face-to-face interactions.
SL2T is our breakthrough sign language-to-text model powering new features for Deaf and hard of hearing users on @Android.
Starting with American Sign Language-to-English on Pixel 11, people can sign directly into Gboard and Live Transcribe instead of typing. pic.twitter.com/p9Vx7tJtLT
— Google DeepMind (@GoogleDeepMind) August 12, 2026
The new feature, powered by an AI model known as sign-language-to-text (SL2T) and developed by Google DeepMind, represents a significant advancement in the translation of sign language. Upon capturing a signing user through the Pixel Camera, the model processes the signs and Gboard promptly converts them into text. This technology can be utilized in various digital environments, from web searches to messaging applications, making it highly versatile.
Upon launch, the feature is designed to translate American Sign Language into English on the Pixel 11 series, with plans for additional languages and extended device support anticipated in the future.
Built with Privacy and Real-World Signing in Mind
Privacy considerations are paramount in this communication tool. The system functionally tracks location points on the user’s body instead of analyzing raw video. This approach ensures that only those coordinates are transferred to Google’s servers for translation, while the original video feed is discarded, thus maintaining user privacy. Google DeepMind has trained the model on over 100,000 hours of data across more than 50 sign languages, ensuring broad applicability.
Interestingly, the model has been specifically tested for left-handed signers, who constitute approximately 10 percent of users, and for one-handed signing, which is common when holding a phone with the other hand. The developers assert that the model outperformed previous systems on standard sign language translation benchmarks, although detailed performance data in varying real-life conditions remain to be seen.
The challenge of translating sign language lies in its unique grammar and vocabulary, which differ significantly from spoken language’s word-for-word translation. According to estimates, around 70 million deaf and hard-of-hearing people worldwide utilize sign language. Google DeepMind posits that this launch marks the first occasion when sign language AI has been integrated into a consumer product at such a broad scale. Ultimately, the efficacy of SL2T in daily signing will determine if it can transcend being just another AI feature and become a practical communication tool for millions.
For further reading on this topic, click Here.
Image Credit: www.digitaltrends.com
