In a groundbreaking development, scientists have successfully helped a woman speak again after 18 years of silence, using an experimental brain-computer interface (BCI) that translates thoughts into spoken words in real time.
The subject of the study, a 47-year-old woman with quadriplegia caused by a stroke, had been unable to speak for nearly two decades. But thanks to a team of researchers in California, she is now regaining her voice, literally. The findings were published Monday in Nature Neuroscience.
The experimental device, implanted during surgery as part of a clinical trial, decodes neural signals in the brain’s speech center and transforms them into verbalized sentences almost instantly.
“It converts her intent to speak into fluent sentences,” said Gopala Anumanchipalli, a co-author of the study and a researcher at the University of California, Berkeley.
Unlike other BCIs that experience delays between thought and speech output, the new system offers near-instantaneous verbalization, an innovation that scientists say significantly enhances the natural flow of conversation.
“This is a pretty big advance in our field,” noted Jonathan Brumberg of the Speech and Applied Neuroscience Lab at the University of Kansas, who was not involved in the study.
To build the system, scientists recorded the woman’s brain activity as she silently attempted to speak specific sentences. Using a set of pre-injury voice recordings, the researchers developed a customized synthesizer to recreate her natural voice. An artificial intelligence model was then trained to translate brain signals into phonemes, basic sound units of speech.
“It’s not waiting for a sentence to finish,” Anumanchipalli explained. “It’s processing it on the fly.” He described it as a “streaming approach,” where each 80-millisecond segment—roughly half a syllable—is immediately fed into a recorder.
The implant is positioned directly on the brain’s speech center, where it continuously picks up neural signals that are decoded and assembled into spoken words. The system functions much like real-time transcription tools used for meetings and phone calls but tailored to the unique neural patterns of the individual.
According to Brumberg, decoding speech at such rapid speeds is essential for preserving the natural rhythm and pace of conversation. The incorporation of voice samples from the individual also enhances the authenticity of the spoken output. “That would be a significant advance in the naturalness of speech,” he said.
The project received partial funding from the National Institutes of Health (NIH), and while recent NIH budget cuts have affected many research initiatives, Anumanchipalli confirmed this study was not among them.
He emphasized that while the device is still in early experimental stages, with “sustained investments,” the technology could be available to patients within a decade.
The success of this clinical trial not only marks a hopeful breakthrough for those suffering from speech loss due to injury or illness, it also opens up transformative possibilities in human communication powered by neuroscience and AI.
Paraluman P. Funtanilla
Paraluman P. Funtanilla is Tutubi News Magazine's Marketing Specialist and is a Contributing Editor. She finished her degree in Communication Arts in De La Salle Lipa. She has worked as a Digital Marketer for start-up businesses and small business spaces for the past two years. She has earned certificates from Coursera on Brand Management: Aligning Business Brand and Behavior and Viral Marketing and How to Craft Contagious Content. She also worked with Asia Express Romania TV Show.
- Paraluman P. Funtanilla#molongui-disabled-link
- Paraluman P. Funtanilla#molongui-disabled-link
- Paraluman P. Funtanilla#molongui-disabled-link
- Paraluman P. Funtanilla#molongui-disabled-link
