Three research teams presented advances in implantable brain-computer interfaces at the Society for Neuroscience meeting in Washington, D.C., that could restore natural communication for people with paralysis.
The findings, reported by The Scientist, center on making systems mobile, keeping them usable as patients lose all muscle control, and recreating the pitch and rhythm of a human voice.
Chethan Pandarinath, a neuroscientist at the Georgia Institute of Technology and Emory University, said current systems rely on bulky computers and suffer from “slow and variable response times” that “impede natural communication.” He added that the need for large computers “really restricts them from being mobile, on-the-go, always available systems.”
His team built an adaptor that converts neural signals into speech-like signals a smartphone’s built-in dictation engine can recognize. Trained on data from two BrainGate 2 trial participants, the phone-based model generated responses faster and with less memory than computer-based systems.
Pandarinath said the platform is now being evaluated in real time; one participant told researchers that when she orders food at a restaurant, “people have to try to read her lips, which can be difficult,” and she was excited the technology could give her another option.
Daniel Rubin, a neurologist at Mass General Brigham and Harvard Medical School, described a patient with ALS who could still move his eyes when a BCI was implanted. The system let him spell words by imagining moving different body parts. Over the following year the patient lost eye movement entirely, leaving the implanted BCI as his only communication channel.
“We quickly set out to make sure that he could use this system anytime he needed, and importantly, that he could use this system when our team wasn’t present,” Rubin said.
After improving their decoding algorithms, the team reanalyzed previously collected neural data and raised the intelligibility of the generated speech to 94%. The outlet notes, however, that “the improved performance has not yet established that the system can achieve the same level of intelligibility during real-time conversations.”
In the same study, a woman with ALS who was once a singer used the system to modulate pitch and produce a vocal scale, showing the technology can capture prosodic features such as pitch, timbre, intonation and rhythm that text-only systems lose. Researchers said additional tests are needed to determine whether similar performance occurs in live speech and whether the approach works across varying levels of impairment.
Chethan Pandarinath earned a Ph.D. from Cornell University in 2010 and a B.S. from NC State University in 2002. He leads the Systems Neural Engineering Lab at Emory University and Georgia Tech, where his group develops deep learning methods to decode neural population dynamics. In 2025, a Cell study from his lab demonstrated inner speech signals in human motor cortex, and he was elected to the AIMBE College of Fellows.
“There’s no cure for ALS, nor will the technology restore your ability to speak like you used to,” the Tech Times UK report states, “but the results suggest a future in which brain-computer interfaces can enable ALS patients to speak more naturally and regain some of the rich expression often lost when information is transmitted via text.”
Related reading
- Study Links Smartphone Era to Sharp Rise in Young Adults’ Intolerance of Uncertainty
- Imsil County Orders Emergency Measures as Tomato Virus Hits 9 Farms
- RBA's Fifth Hike Adds About A$94 a Month on A$600,000; Three Voted to Hold (daybreakwire.com)
- US Natural Gas Futures Jump 5.11% on Cold Weather Outlook, Short Covering (news-usa.today)