Google is developing a new beta app, Project Relate, that aims to provide a voice assistant to people with speech impairments. An AI-powered synthesizer will transcribe speech in real time as well as synthesize what the person is saying. Julie Cattiau, product manager at Google’s AI division, explains that this will work in the following way: early testers of the project will be asked to record a set of phrases that will teach the app how to better understand the user’s unique speech patterns.
This app is part of a wider endeavor, Project Euphoria, that began in 2019 with the collection of data to improve Google’s Al speech-related algorithms for people who “have difficulty being understood by others,” such as those affected by neurological conditions.
The Repeat feature will restate what the user is saying in a “clear synthesized voice,” which Google hopes will aid face-to-face conversations and help people with speech impairments issue a command to smart home devices. The Assistant feature will allow users to speak to the Google Assistant from directly within the Relate app.