Revolutionary technology helps man with chronic neurological disease speak again

A man with a debilitating neurological disorder has been able to “speak” again thanks to a computer that translates brain activity into text that is read aloud.

The US UC Davis Medical Center has developed a new brain-computer interface (BCI) that translates brain signals into speech with 97% accuracy, making it the most accurate system of its kind.

In the experiment, scientists implanted sensors in the brain of a man with severe speech impairment due to amyotrophic lateral sclerosis (ALS), which affects the nerve cells that control movement throughout the body. The disease causes a gradual loss of the ability to stand, walk and use the hands, and can also cause a person to lose control of the muscles used for speech. The results revealed that the patient, Casey Harrell, 45, was able to communicate his intended speech within minutes of activating the system.

The new technology can interpret brain signals when the user tries to speak, and convert them into "audio text" by a computer.

"Our BCI technology helped a paralyzed man communicate with friends, family and caregivers," said neurosurgeon David Brandman, from the UC Davis Center.

The developed system was able to decode Harrell's speech in real time, with continuous system updates to keep it working accurately. The decoded words were displayed on the screen. The recordings were read aloud in a voice that sounded like Harrell's voice before he was diagnosed with ALS, as the scientists used a program that was trained on existing audio samples of Harrell's voice before he was diagnosed with ALS.

In the first speech data training session, the system took 30 minutes to achieve 99.6% accuracy in the resulting words, with a vocabulary of 50 words.

In the second session, the potential vocabulary increased to 125,000 words. With just 1.4 additional hours of training data, the BCI achieved 90.2% accuracy in words with this greatly expanded vocabulary. After continued data collection, the BCI maintained 97.5% accuracy.

“At this point, we can correctly decode what Casey is trying to say 97% of the time, which is better than many commercially available smartphone apps that try to interpret a person’s voice,” Brandman said. “This technology offers hope to people who want to talk but can’t. I hope it will help patients in the future talk to their family and friends.”

The study was published in the New England Journal of Medicine.

Source: Medical Express - Published on 16-August-2024 https://ar.rt.com/y44g