A man who is severely paralyzed was able to communicate successfully using cutting-edge technology that translates signals from his brain to his vocal tract into words that are displayed on a screen.
Researchers at UC San Francisco developed this technique that allows people with speech loss to communicate more naturally than other methods that we’ve seen so far. Previous technology enabled paralyzed users to communicate, but only by typing out one letter at a time, which is a painstakingly slow process that requires the parts of the brain that control the arm or hand.
On the other hand, The USCF team developed an implant that can be placed directly on the part of the brain that’s dedicated to speaking. This way, the subject can mentally activate the brain patterns they would normally use to pronounce a word, and the system can take those brain patterns and turn them into words in their entirety rather than single letters.
To create the implant, volunteer patients with normal speech let the team to analyze their brain recordings for speech-related activities. The researchers were then able to study those patterns and develop new methods to decode them in real-time. Once the team was ready to see if brain signals controlling the vocal tract would still be intact in patients who have been paralyzed for a long stretch of time, an anonymous participant who goes by Bravo1 worked with researchers to create a 50-word vocabulary that the team could interpret through advanced computer algorithms.
Patient Bravo1 was then asked to reply to simple questions such as “How are you today?” and “Would you like some water?” The patient would then attempt to speak, but instead, the reply was displayed on a screen as “I am very good,” and “No, I am not thirsty.”
The system is able to decode the patient’s speech at up to 18 words per minute with up to 93 percent accuracy and a 75 percent median accuracy, which is a significant improvement on previous neuroprosthetic systems.
The team is excited about the promising results. MD and Chair of Neurological Surgery at UCSF and senior author on the study Edward Chang says that “it shows strong promise to restore communication by tapping into the brain’s natural speech machinery.”
The next step for the researchers is to expand the trial so that more participants can participate, and to amplify the vocabulary that the system can recognize.
