(SACRAMENTO, Calif.) – A groundbreaking brain-computer interface (BCI) developed at UC Davis Health is translating brain signals into speech with an unprecedented accuracy rate of up to 97%, setting a new standard for neuroprosthetic technology.
The innovative system was implanted in the brain of a man with severely impaired speech due to amyotrophic lateral sclerosis (ALS), a progressive neurological disorder. Remarkably, the patient could communicate his thoughts within minutes of activating the device. This cutting-edge research was recently published in the New England Journal of Medicine.
Giving Voice to Those Silenced by ALS
ALS, also known as Lou Gehrig’s disease, affects motor neurons that control movement, leading to gradual paralysis. Over time, individuals with ALS lose the ability to walk, use their hands, and even speak clearly. For many, this loss of communication is one of the most isolating aspects of the disease.
The newly developed BCI technology offers a breakthrough solution. It deciphers brain signals generated when the user attempts to speak and converts them into text, which the system then “speaks” aloud through a computer.
“This technology has empowered a man with paralysis to reconnect with loved ones and caregivers,” said Dr. David Brandman, a neurosurgeon at UC Davis and co-principal investigator of the study. “Our findings represent the most accurate speech neuroprosthesis ever reported.”
How the BCI Works
When a user attempts to speak, the BCI captures brain activity and translates it into text displayed on a computer screen. The text is then vocalized using synthesized speech.
To test the system, researchers enrolled Casey Harrell, a 45-year-old man with ALS, in the BrainGate clinical trial. Harrell had severe motor impairment and slurred speech, requiring others to interpret his words. In July 2023, Dr. Brandman surgically implanted four microelectrode arrays into the left precentral gyrus of Harrell’s brain, a region responsible for coordinating speech. These arrays recorded brain activity from 256 cortical electrodes.
“We’re essentially decoding the brain’s attempts to move muscles and produce speech,” explained Dr. Sergey Stavisky, a neuroscientist and co-principal investigator. “By capturing patterns of brain activity, we can translate them into phonemes—basic speech units—and then reconstruct the words the individual is trying to say.”
Faster Learning, Greater Accuracy
Traditional BCI systems often struggle with slow learning curves and high error rates, limiting their effectiveness. This new system, however, sets a higher standard.
“Previous systems frequently made errors, making consistent communication challenging,” said Brandman. “Our goal was to develop a reliable tool that allows users to express themselves clearly and effortlessly.”
In the initial training session, Harrell’s system reached 99.6% word accuracy within 30 minutes using a 50-word vocabulary. Later sessions expanded the vocabulary to 125,000 words. After just 1.4 additional training hours, the system achieved 90.2% accuracy, and with ongoing refinement, maintained an impressive 97.5% accuracy rate.
In real-time conversations, the decoded text appeared on a screen and was spoken aloud using a voice synthesized from Harrell’s pre-ALS recordings, adding a deeply personal touch.
“The first time Casey used the system, he cried with joy when the words he was trying to say appeared accurately on the screen,” recalled Stavisky. “It was an emotional moment for all of us.”
A Life-Changing Technology
Over 32 weeks, Harrell participated in 84 data collection sessions, totaling 248 hours of BCI-assisted communication, including video chats and in-person conversations.
“Losing the ability to communicate feels like being trapped,” Harrell shared. “This technology gives people like me a way back to life and society.”
The study’s lead author, Dr. Nicholas Card, highlighted the importance of early clinical trials. “Casey and others like him are extraordinary,” he said. “Their courage and dedication are driving the development of a system that will restore communication and mobility for countless individuals in the future.”
As the field of brain-computer interfaces advances, this groundbreaking technology offers new hope for individuals living with ALS and other conditions that impair speech, providing a lifeline to reconnect with the world.
