Scientists from Duke University have come up with a brain implant with an impressive 84% accuracy in translating thoughts into speech.
The flexible device the size of a postage stamp, and equipped with 256 tiny brain sensors, aims to help those with severe neurological issues, including ALS and locked-in syndrome, communicate more effectively.
Gregory Cohan, one of the professors who helped develop the device, explained that many patients suffer from these disorders that impair their ability to speak.
"The current tools available to allow them to communicate are generally very slow and cumbersome," he added.
How Did the Device Perform During Testing?
The implant was initially tested on four patients undergoing brain surgery for conditions like Parkinson's disease.
Inserted within a tight 15-minute window, it recorded brain signals to understand mouth movements and sound pronunciations. A quick "listen-and-repeat" activity enabled the AI system to mimic the user's speech patterns.
“I like to compare it to a NASCAR pit crew," Cohan explained. “We don't want to add any extra time to the operating procedure, so we had to be in and out within 15 minutes."
Results showed an average of 40% translation accuracy, with some sounds reaching an impressive 84%.
While similar devices exist for speech-impaired individuals, they often require a secondary device and an electrical outlet.
Duke University's innovation aims to create a cordless version, enhancing mobility during conversations. A $2.4 million grant from the National Institutes of Health supports this effort.
While the device shows promise in becoming a commercial tool in the future, concerns over ethical and safety issues accompany this brain technology.
In light of this, the Duke University team is addressing worries about long-term effects, patient feedback, and operational challenges.
The recent developments also come a few days after Elon Musk's Neuralink began its search for the ideal patient to partake in the human trials for a brain implant.
The implant would then acquire data from the premotor cortex to translate human thoughts into discernible computer commands.
Musk envisions the technology enabling paralyzed individuals to control smartphones with their minds.