People unable to speak due to brain damage or injury may be able to speak again, thanks to a revolutionary brain implant that creates words directly from a person’s thoughts.
Researchers at the University of California, San Francisco, say they have successfully reconstructed “synthetic” speech from the implant, scanning brain signals of volunteers as they read hundreds of sentences out loud.
The tech is still in early development, but shows the potential of translating thoughts of mute people in real-time.
While the experiment was conducted only with people who could speak, the team found that speech could be synthesized from participants even when they only mimed the sentences.
“Very few of us have any idea of what’s going in our mouths when we speak,” said Edward Chang, lead study author.
“The brain translates those thoughts into movements of the vocal tract and that’s what we’re trying to decode.”
This paves the way for an implant that translates words from the brain activity of patients who know to speak, but have simply lost the ability to do so.
Gopala Anumanchipalli, co-author of the study, says that the words tested so far would be added to database from which users could one day build more complicated statements; at the moment, only simplistic statements can be constructed.
“We used sentences that are particularly geared towards covering all of the phonetic contexts of the English language,” he said. “But they are only learned so they can be generalized from.”
Chang said this had potential to act as a starting point for patients re-learning to talk after injury, who could train to control their own simulated voice from the patterns learned from able speakers.
Chiming in from a linked comment piece, Chethan Pandarinath and Yahia Ali, from the Georgia Institute of Technology, Atlanta, call the study “compelling”.
“With continued progress, we can hope that individuals with speech impairments will regain the ability to freely speak their mings and reconnect with the world around them,” they wrote.
Photo Credit: Jolygon/Shutterstock.com; Popartic/Shutterstock.com