A University of Melbourne startup called Fluent is building a brain-computer interface that sits just under the scalp and decodes the brain’s electrical signals to help people with speech impairments communicate — no surgery, no buttons required.
For millions of people living with neurological conditions like motor neurone disease or multiple sclerosis, the ability to speak can gradually disappear. Most assistive technologies on the market today require users to laboriously select letters or words on a screen using their eyes or fingers — a slow and exhausting process. A University of Melbourne spinout company called Fluent wants to change that with a brain-computer interface that reads thought-driven brain signals and converts them into text or audio.
How the Technology Works
The device is designed to sit beneath the scalp but outside the skull, threading a minimally invasive path between existing, far more surgical alternatives. It targets a specific region of the brain responsible for coordinating the muscle movements involved in speech.
“Our device will be positioned above an area of the brain called the motor cortex which controls speech muscles,” Tim Mahoney, Fluent co-founder and biomedical engineer, said in a news release.
Mahoney uses an accessible analogy to explain how the system captures and interprets neural activity. Every distinct mouth or jaw movement triggers a unique electrical pattern in the motor cortex — patterns he likens to QR codes. Critically, those same patterns fire even when someone with a speech impairment tries but fails to produce sound.
“These ‘QR codes’ also occur when a person with impaired speech is attempting to speak. Our device can capture these codes in a sequence, which tells us what someone is trying to say,” he said.
A machine learning model is being built to translate those recorded patterns into readable text or audible speech, allowing non-verbal individuals to express themselves without pressing a single button or uttering a single word.
What the Research Has Shown So Far
Early human testing took place inside electromagnetically shielded lab rooms at the newly opened Aikenhead Centre for Medical Discovery at St. Vincent’s Hospital Melbourne. In those sessions, participants wore 144 scalp electrodes that tracked motor cortex activity as they spoke aloud, mimed speech silently, and mentally imagined saying different phrases.
“After building the largest English dataset of its kind, we partnered with a Japanese team leveraging an even larger dataset. This collaboration proved that a model can accurately isolate the correct phrase from a pool of 128 options with 96 per cent accuracy,” Mahoney added.
That level of accuracy — selecting the right phrase from 128 possibilities nearly 96% of the time — marks a substantial leap from what communication boards and eye-tracking tools can offer. During his doctoral research in accessible brain-computer interfacing, Mahoney also demonstrated that signal quality is comparable whether the electrodes sit below or above the scalp, which means the team can iterate and improve the technology without waiting on complex surgical protocols.
Why It Matters
The gap between what existing brain-computer interfaces can do and what patients can actually access has long been defined by one word: surgery. Devices that require electrodes implanted inside the skull carry significant risk, expense, and recovery time — barriers that put the technology out of reach for most people who could benefit from it.
“Until now, this technological capability was only thought possible using highly invasive electrodes implanted inside the skull. With a safety profile that’s even better than a routine cochlear implant, the technology will be more accessible to the broader population,” added Mahoney.
By lowering that barrier, Fluent’s approach could extend life-changing communication tools to a far wider population of patients — including younger people recently diagnosed with progressive neurological conditions who want to maintain independence as long as possible.
Clinical studies using Fluent’s insertable electrodes are scheduled for later this year, a milestone that will move the technology from controlled lab settings into real-world testing.
University Support and Funding
Fluent emerged from the University of Melbourne’s broader innovation infrastructure, receiving early-stage proof-of-concept funding and support through the Melbourne Entrepreneurial Centre’s accelerator program before attracting outside investment.
“Fluent was born out of the University’s innovation ecosystem with its researchers securing Proof of Concept funding, nurtured by the Melbourne Entrepreneurial Centre’s accelerator program and now securing investment to develop human clinical trials,” Mark Cassidy AM, deputy vice-chancellor (research) professor at the University of Melbourne, said in the news release.
The company has now raised more than $2 million from a diverse group of backers, including the University of Melbourne Genesis Pre-Seed Fund, Galileo Ventures, Multiple Sclerosis Western Australia, and international investors from New York, London and Auckland.
“This startup is working to improve the independence of people living with constraining illnesses and transforming lives as a result,” Cassidy added.
Source: University of Melbourne