The brain signals of a 46-year-old man who lost his voice to ALS were decoded by experimental brain implants and translated directly into intelligible speech. A 58-year-old woman, a trained singer, whose voice was gradually silenced by the disease, was able to sing a musical range.
New advances from University of California, Davis researchers, to be presented next week at the Society for Neuroscience conference in Washington, demonstrate that research on brain-computer interfaces is moving from proof-of-concept demonstrations to everyday utility.
“This system, for the first time, performs quite well,” said Cindy Chestek, a biomedical engineer at the University of Michigan, who was not involved in the new work. “Good enough for real communication. It’s good enough for singing.”
For years, brain implants combined with increasingly powerful AI have been shown in remarkable case studies to restore a person’s ability to communicate, ask questions, and even make jokes. The sense of connection the technology provided for patients—who had lost control of the muscles that allowed them to speak, due to ALS or a stroke—could be profound, but it was also disrupted, as the systems first translated signals into text, then into speech.
A person would attempt to speak and electrodes surgically implanted in the brain would detect electrical signals. The software would decode these signals into phonemes, the sound building blocks of words. Participants watched words appear one by one on a screen, thus constructing a sentence. They would then trigger a cursor click to signal that they were done. Then a computer-generated version of their voice would speak.
Research subjects could convey their thoughts in great detail, but with a slight delay, making it difficult to truly join the conversations. People often communicate with the rhythm, timing, and emphasis of their speech, not by composing complete responses and mentally pressing “send.”
“Imagine if you can only talk by texting for the rest of your life, that’s not enough, is it?” said Maitreyee Wairagkar, a neuroengineer at the Neuroprosthetics Laboratory at the University of California, Davis, who led the work on the two patients.
One of them, Casey Harrell, a climate activist who lost his voice to ALS, began participating in the study at UC Davis in 2023, when four arrays of 256 electrodes were surgically implanted in his brain. He was 45 years old. He used the system to generate nearly 2 million words in two years, including conversations with family, messages to friends, writing emails and browsing the Internet on his laptop for work.
In 2025, the team published in Nature a remarkable feat of generating expressive speech in real time – direct, from brain to voice – with a delay of just 10 milliseconds.
The feat was impressive, but had limitations. Those listening could only accurately transcribe a little more than half of his words.
In the new work, described in a preprint, the researchers refined the algorithms and showed that by reanalyzing the same data set, they could increase word intelligibility by up to 94%.
“I think this is a really clinically meaningful improvement and not just impressive machine learning,” said Christian Herff, who directs the Neural Interface Lab at Maastricht University in the Netherlands and was not involved in the study.
Mariska Vansteensel, a brain-computer interface researcher at University Medical Center Utrecht in the Netherlands, applauded the work, but said that because it was testing the technology on already collected data, it would be important to see how well it worked in real time and in people with varying degrees of speech loss.
She added that for such systems to be useful outside of a research setting, they will need to be fully implantable, unlike systems that currently require participants to be connected to a computer through ports in their skull. Companies such as Neuralink and Paradromics are developing and testing wireless systems, but how these technologies work remains to be demonstrated.
In a second experiment presented by the same team, a participant with ALS who could no longer speak intelligibly when she enrolled in the study had 384 electrodes implanted in her brain. This patient, known as T21, tried to sing and modulate the tone of her voice, controlling pitch nuances that can mean the difference between asking a question and saying something happy or angry.
As the speech generation capabilities of these systems have improved, research teams have also begun to focus on other essential facets of human communication, ranging from gestures to tone.
“They are not only for essential communication, they are also expressive; you can show more of your personality,” said Dr Wairagkar. This research is ongoing and the results are preliminary and unpublished.
With this progress, participant T21 was not only able to recreate the musical qualities of a conversation, using a questioning or happy tone, but she could also sing a scale. The experience was so profound, Dr Wairagkar recalls, that afterwards the former singer cried tears of joy.
Gn tech
