Credit: Shutterstock
A new device may remove the communication barrier for people with impaired speech.
The device comes from Fluent, a startup founded at the University of Melbourne. Fluent is developing a brain-computer interface for people with conditions such as Motor Neuron Disease (MND) and Multiple Sclerosis (MS). These conditions can damage the nerves and muscles involved in speech, leaving people unable to speak even though their minds remain fully active.
The Problem With Current Devices
People who lose their speech today often rely on communication boards or eye-tracking screens. These tools let a person select letters or words one at a time. The process is slow and can be frustrating for both the user and the people trying to communicate with them.
Brain-computer interfaces have offered a faster alternative in recent years. However, most existing systems require electrodes to be surgically implanted inside the skull. This surgery carries risks and is expensive, which limits how many people can access the technology.
How Fluent’s Device Is Different
Fluent’s device works differently from other brain-computer interfaces. Instead of being implanted inside the skull, it is placed just under the scalp, avoiding the need for major brain surgery. It remains above the motor cortex, the part of the brain that controls the muscles used for speech.
Whenever a person speaks or even tries to speak, their motor cortex produces distinct patterns of electrical signals for each mouth and jaw movement. Fluent’s device captures these signals, and an AI system analyzes them to determine what the person is trying to say. The end goal is to convert this brain activity directly into text or audio.
Dr. Tim Mahoney, the company’s co-founder and a biomedical engineer, showed during his PhD research that signal quality from this less-invasive under-scalp placement was nearly as clear as signals recorded from electrodes placed beneath the skull. This finding matters because it means the technology can move forward without requiring more complex surgical procedures.
Testing So Far
In initial human trials at St Vincent’s Hospital Melbourne, volunteers wore a cap with 144 electrodes while they spoke, silently mouthed words, or imagined speaking different phrases. The electrodes recorded brain signals associated with speech.
Using the data collected, Fluent built one of the largest English-language datasets of its kind for this purpose. The team then partnered with a Japanese research group that had an even larger dataset. Together, they showed that their model could correctly identify the intended phrase from 128 possible options with 96% accuracy.
Fluent plans to begin clinical trials of its implantable device later this year. According to co-founder Dr. Tim Mahoney, the procedure is expected to be safer than a standard Cochlear Implant, making the technology more practical and accessible than current brain-computer interfaces.
Funding and Support
Fluent has raised more than $2 million so far. Investors include the University of Melbourne Genesis Pre-Seed Fund, Galileo Ventures, Multiple Sclerosis Western Australia, and international backers such as Jumpspace Ventures in New York, Founder’s Factory in London, and Pacific Channel in Auckland.
Why This Matters
This technology could make speech-restoring brain implants much more accessible. Most existing systems, including those developed by companies like Neuralink and Synchron, require electrodes to be implanted inside the brain through major surgery. Fluent’s device aims to provide similar results while using a much less invasive procedure, making it safer, more affordable, and accessible to many more people with speech impairments.
However, as with all brain-computer interfaces, there are still important questions about privacy, security of brain data, and patient consent, especially for people with severe disabilities. Researchers will need to address these issues as the technology moves closer to everyday use.
