UC researchers make breakthrough on brain-to-speech device
There has been a recent breakthrough in the development of a brain to voice neuroprosthesis. These studies have been conducted by UC Berkeley and UC San Francisco. This new technology aids in restoring naturalistic speech to people with paralysis. Originally, the intent of this research was to improve the latency of the brain-to-voice system, but they have now made it possible to synthesize brain signals into speech in close to real-time. This implant sits on the part of the personal brain that is responsible for producing speech.
The vocal track will be able to be done in the patient’s voice using trained AI models by using the part of the patient’s brain that is still intact. The technology that was used and created was inspired by Automatic Speech Recognition that are used in softwares such as Siri to achieve a streamable algorithm.
For the future, the team is working on making the device wireless and portable, including expressivity control, and making the system more closed loop. While there are still many challenges to the work, the team will remain working towards important steps to help people who have lost their ability to speak.
