A.L.S. Stole His Voice. A.I. Retrieved It

Last month, Casey Harrell’s story of regained communication using our brain-computer interface made headlines. Today, we want to spotlight the brilliant minds who made such a feat possible.

The groundbreaking study, published in The New England Journal of Medicine, was led by:

• Dr. Nicholas Card, Neuroscientist, University of California, Davis
• Dr. David Brandman, MD, PhD, Neurosurgeon, University of California, Davis
• Dr. Sergey Stavisky, Neuroscientist, University of California, Davis
• Dr. Leigh H., Neurologist, Brown University & Dept. of Veterans Affairs

Their approach uniquely combined high-density intracortical recording with rapid calibration and AI-powered decoding, setting a new standard in BCI performance.

Key innovations:

1.) Four Utah Arrays were implanted in the “speechiest parts of the brain,” as Dr. Stavisky puts it for The New York Times. This includes area 55b and three areas in the ventral precentral gyrus associated with speech production.

2.) 256 intracortical recording sites from our NeuroPort Arrays (3.2mm x 3.2mm, inserted 1.5mm into cortex).

3.) Advanced algorithms translated neural signals into speech with unprecedented accuracy. The system uses a neural network to predict phonemes every 80ms, followed by open-source language models to translate these into words and sentences.

The results speak for themselves:

• 99.6% accuracy with a 50-word vocabulary on the first day of use, after only 30 minutes of training data
• 90% accuracy with 125,000 words by day two, after just 1.4 additional hours of system training
• 97.5% sustained accuracy over eight months, with the participant communicating at a rate of approximately 32 words per minute

Casey used the neuroprosthesis to communicate a total of 22,679 sentences during 72 sessions over 8.4 months (that’s 248.3 cumulative hours), demonstrating both the system’s durability and its practical applicability in daily life. As Dr. Brandman shared, “It went from a scientific demonstration to a system that Casey can use every day to speak with family and friends.”

This research builds on years of work in the BCI field, including contributions from Dr. Edward Chang at UCSF, whose team developed different speech implants.

At Blackrock Neurotech, we’re proud to provide the neural interface technology that enables such groundbreaking research. We’re committed to pushing the boundaries of what’s possible in neurotechnology, always with the goal of improving lives.

A huge thank you to all the researchers, clinicians, and most importantly, BCI Pioneers like Casey who make this work possible. Together, we’re writing the future of neurological care.

You can read more about Casey’s story here.