People who have lost the ability to speak due to a stroke, cerebral palsy, Charcot’s disease, Alzheimer’s, brain damage, cancer, etc. may soon regain their voice. Researchers at the University of California, San Francisco (UCSF) announced on Wednesday that they have successfully translated brain activity into words, moving “from thought to speech”.
“Very few of us have any real idea, actually, of what’s going on in our mouth when we speak,” said neurosurgeon Edward Chang, lead author of the study. “The brain translates those thoughts of what you want to say into movements of the vocal tract, and that’s what we’re trying to decode”.
In the study published in the journal Nature, the scientists explain that they implanted electrodes in the brains of five volunteer patients with epilepsy. They tried to decode the signals from a special area of the brain “dedicated to controlling the organs needed to speak: the tongue, lips, larynx and cheeks”. From these “electrical signals,” an artificial intelligence “imagines the ‘movements’ that these various organs would carry out. Then a second algorithm transforms these movements into synthetic words”. The patients read aloud several hundred sentences. They also “mimicked words without producing sound”.
“When we first heard the results, we couldn’t believe our ears. It was incredibly exciting that a lot of the aspects of real speech were present in the output from the synthesizers,” said Josh Chartier, co-author of the study and a PhD student at the UCSF. “Clearly, there is more work to get this to be more natural and intelligible but we were very impressed”.
Some people use devices that track eye or residual facial muscle movements to laboriously spell out words letter-by-letter, but producing text or synthesized speech this way is slow, typically no more than 10 words per minute. Natural speech is usually 100 to 150 words per minute, a fluidity that the device would like to achieve.
For the time being, things are still at the proof-of-concept stage and “it will still take a lot of experiments to make this technology effective”. Future studies are expected to include patients who are unable to speak.