A joint research team from the University of California, San Francisco (UCSF) and the University of California, Berkeley (UC Berkeley) develops a brain-computer interface (BCI) device that translates the thoughts of paralyzed patients who have lost their language abilities into real-time speech. The photo shows the experiment scene from 2023. The research team shortens the speech conversion delay time from several seconds to milliseconds (㎳) during this time./Courtesy of Noah Berger
A joint research team from the University of California, San Francisco (UCSF) and the University of California, Berkeley (UC Berkeley) develops a brain-computer interface (BCI) device that translates the thoughts of paralyzed patients who have lost their language abilities into real-time speech. The photo shows the experiment scene from 2023. The research team shortens the speech conversion delay time from several seconds to milliseconds (㎳) during this time./Courtesy of Noah Berger

A woman who lost the ability to move her limbs and speak due to a stroke has found her voice again after 20 years, thanks to a device that reads her thoughts in real time and is implanted in her brain. Although similar experiments have succeeded before, there was a delay in converting thoughts into speech. This time, it is expected to significantly contribute to helping patients with language disabilities regain their daily lives by effectively realizing thoughts into speech in real time.

A joint research team led by Edward Chang, a professor at the University of California, San Francisco (UCSF), and Gopala Anumanchipalli, a professor at the University of California, Berkeley (UC Berkeley), announced on the 1st that they have successfully expressed the thoughts of a female patient paralyzed by a stroke into speech by implanting a brain-computer interface (BCI). The research results were published in the international journal "Nature Neuroscience" on the same day.

BCI is a technology that connects the brain and computer to restore bodily functions or convey human intentions to external devices. Electrodes implanted in the brains of paralyzed patients capture thoughts and manipulate robots or computers, converting them into text or speech.

The clinical trial involved a woman who, at 30 years old in 2005, lost the ability to move her limbs and speak due to a stroke. The BCI device succeeded in real-time speech implementation by reducing voice conversion delay to the level of a few milliseconds (1 ms is one-thousandth of a second).

Previously, the research team had developed the device that reads brain signals in 2023, but there was a delay of several seconds between thought and speech conversion, making practical use difficult. Even a delay of one second can hinder natural conversation when people talk. At that time, they adopted a method to enhance communication by implementing speech with a digital avatar.

The research team connects the device implanted in the brain of the paralyzed patient to a computer that synthesizes speech. This time, the brain signals that arise when the patient thinks about reading a sentence are synthesized into real-time speech./Courtesy of Noah Berger

The research team presented sentences to the woman and had her think of speaking before measuring her brain signals. The data learning included sentences containing about 1,024 words. The BCI device captured signals from the motor cortex, the brain area responsible for language ability, and converted the woman's thoughts into speech.

The computer-generated voice was recreated based on recordings of the woman’s voice before she suffered the stroke. As a result, she successfully thought of sentences and read them out loud in 80 milliseconds. Even words not used in training were converted into speech normally.

The research team noted, "The real-time voice generation system has the potential to maintain the flow of natural conversation," adding, "We hope this research will help patients who have lost their ability to communicate regain their communication skills."

However, since there was only one participant in this study, additional research is needed to confirm the performance and side effects of the BCI device. The research team stated, "Further studies involving more patients are necessary," while adding, "If patients who have lost their language ability can converse naturally and smoothly, their quality of life will greatly improve."

References

Nature Neuroscience (2025), DOI: https://doi.org/10.1038/s41593-025-01905-6

Nature (2023). DOI: https://doi.org/10.1038/s41586-023-06443-4