As in year’s past, the Google I/O 2019 keynote has taken time to focus on Google’s latest accessibility tech. One of this year’s demos, Project Euphonia, is an impressive project designed to let the Google Assistant’s voice recognition understand those with speech impairments.
During last year’s keynote, Google demonstrated a way to help more users communicate, by creating morse code input for Gboard complete with word predictions. This year, Google has created a way for a new segment of users to communicate, called Project Euphonia.
Project Euphonia is designed to personalize speech recognition and other non-verbal communications for those with speech impairments of any kind. Some demonstrations shown include a personalized voice recognition model for a deaf man with slurred speech, and another man who is restricted to typing one letter at a time through eye movements.
For these and many more, Google’s Project Euphonia will make more rapid communication possible, where it once seemed an impossibility, through the Google Assistant. To make this happen though, Google needs to collect as many samples as possible from the speech-impaired of all sorts.
If you or anyone you know has any kind of speech impairment, Google is requesting recordings to help generate a model for Euphonia to understand their specific disability.
Project Euphonia was created with help from Googler, Professor of Applied Mathematics, and Professor of Physics at Harvard University, Michael Brenner.