Meet Google SignGemma: An AI Tool That Can Translate Sign Language Into Spoken Text

Google has introduced a new AI model named SignGemma, which is capable of translating sign language into spoken text. This tool is specifically designed to convert various sign languages into written language. Although it supports multiple languages, its primary focus and testing have been on American Sign Language (ASL) and English.

Currently, SignGemma is in the testing phase, with plans for a public release by the end of the year. The model was first presented at the Google I/O 2025 keynote. During this event, Gus Martins, Gemma's Product Manager, described it as Google's "most capable sign language understanding model ever."

Google's SignGemma

Google DeepMind's official X account (formerly Twitter) shared a demo of the AI model along with a few key details. The upcoming AI tool aims to bridge communication gaps for individuals who rely on sign language. SignGemma's development highlights Google's commitment to enhancing accessibility through technology.

By focusing on ASL and English initially, the company aims to refine the model's accuracy before expanding its capabilities to other languages.

Future Prospects and Launch Plans

The public launch of SignGemma is anticipated by the end of this year. As it undergoes further testing and refinement, Google aims to ensure that the tool meets high standards of accuracy and reliability for users worldwide.

By introducing SignGemma, Google seeks to empower individuals who use sign language by providing them with a reliable translation tool. This initiative underscores the company's dedication to leveraging AI for social good.

The introduction of SignGemma marks a significant step forward in making communication more inclusive. As technology continues to evolve, tools like this have the potential to transform how people interact across different languages and cultures.