Brain Implants Let ALS Patients Speak and Sing Again, UC Davis Team Reports
Researchers at the University of California, Davis say refined software raised the clarity of brain-decoded speech to 94 percent, and a former singer who lost her voice to ALS was able to sing a musical scale.
Two people living with ALS can now speak in clear, expressive voices through experimental brain implants, and one of them, a trained singer, was able to sing a musical scale, researchers at the University of California, Davis report in work to be presented next week at the Society for Neuroscience conference in Washington.
The system reads electrical signals from electrodes implanted in the brain when a person tries to speak. Software turns those signals into phonemes, the sound units that make up words, and a computer voice speaks them aloud in a voice built from recordings the patient made before losing speech.

Casey Harrell, a climate activist who lost his voice to ALS, joined the UC Davis study in 2023, when surgeons implanted four arrays with 256 electrodes in his brain. He has used the system at home for about two years to produce close to 2 million words, in conversations with his family, in emails and in work on his laptop.

From half the words to 94 percent
In 2025, the UC Davis team reported in the journal Nature that it could turn brain signals into speech with a delay of only 10 milliseconds. Listeners could still make out only a bit more than half of the words, which limited the system in everyday conversation.
In the new work, described in a preprint, the researchers tuned the decoding software and reanalyzed the same recordings. Word clarity rose to as high as 94 percent, according to the team. Christian Herff, who leads the Neural Interfacing Lab at Maastricht University and was not involved in the study, said the gain was a real clinical improvement rather than only a machine-learning result.
Cindy Chestek, a biomedical engineer at the University of Michigan who was not involved in the work, said the system had reached a point where it was “good enough for real communication” and “good enough to sing.”
A singer gets her voice back
A second patient in the study, a 58-year-old woman known as T21 who sang opera before ALS took her voice, received 384 electrodes. She can now shape the pitch of her synthesized voice to ask a question, sound joyful or sound angry, and she sang a musical scale. Maitreyee Wairagkar, the neuroengineer at UC Davis who led the work, said the woman cried tears of joy after her scales were restored. The finding is preliminary and unpublished.

The work still has limits. Mariska Vansteensel, a brain-computer interface researcher at University Medical Center Utrecht, said the 94 percent figure came from reanalyzing data already collected, so the team still needs to show the tuned system works as well in real time and in people with different degrees of speech loss. Current systems also need to be fully implantable; the ones in the study connect to a computer through ports in the skull. The implants do not cure ALS or restore natural speech.
Reporting based on coverage by the New York Times, Our Health Needs and Donga Science.