Breakthrough in brain research: The technology that gives a voice to the paralyzed

Researchers at the University of California, San Francisco have developed a brain-computer interface that translates brain signals into speech and body gestures using a virtual avatar. In a preliminary trial, participants with severe paralysis were able to communicate using words and gestures simultaneously, without activating muscles. The system recognized gestures with up to 85% accuracy in conversation. The technology is still in research stages and is not available for routine use.

Researchers at the University of California, San Francisco have developed an innovative system that translates brain activity into speech and body gestures using a virtual avatar. The system, known as a brain-computer interface (BCI), picks up electrical signals from the brain through a tiny array of electrodes implanted surgically in the area responsible for planning and controlling movement. The signals are translated by machine-learning algorithms into digital commands that operate the avatar. The main innovation is the ability to combine speech and body language in real time. A preliminary trial involved three people with severe paralysis. The results showed the ability to identify individual body movements with up to 82% accuracy, and when speech and gestures were combined simultaneously, accuracy stood at about 65%-70%. In conversation tasks, the system recognized gestures with at least 85% accuracy and speech with at least 75% accuracy. The researchers, led by Dr. Edward Chang, stress that the technology is in research and development stages and is not an available treatment for routine use. The study was published in the journal Nature Neuroscience and funded by the National Institutes of Health in the United States. A wireless implanted version is planned for development next.