The goal of Google's Project Euphonia is to help people be better understood. In order to do that, we work on improving automated speech recognition. Our research particularly focuses on how we can train computers and mobile devices to better understand people who have difficulty pronouncing words due to neurologic conditions (e.g., stroke, traumatic brain injuries, cerebral palsy and ALS).
These improvements to speech recognition are only possible if we have speech samples to train the system. If you have a voice that is difficult to understand due to a condition (but not just because of an accent), you can help! Please fill out this form to volunteer and record a set of phrases.