A team of neuroscientists, neurosurgeons, and engineers have successfully created a speech prosthetic that can translate a person’s brain signals into spoken words. This cutting-edge technology holds the potential to help individuals with neurological disorders, who are unable to speak, regain the gift of communication through a brain-computer interface.
While this breakthrough is undoubtedly promising, there is still much progress to be made before the speech prosthetic becomes widely available. Viventi acknowledges that the current decoding rate remains slower than natural speech but highlights the potential for improvement in the near future.
Published in the journal Nature Communications, the speech prosthetic represents a significant advancement in the field. Gregory Cogan, a neurology professor at Duke University’s School of Medicine and one of the lead researchers, emphasised the importance of this innovation for patients suffering from conditions such as amyotrophic lateral sclerosis (ALS) or locked-in syndrome. These individuals often face significant challenges in expressing themselves, and the current communication tools available to them are slow and cumbersome.
The existing speech decoding rate, comparable to listening to an audiobook at half-speed, stands at a mere 78 words per minute, while the average person speaks at approximately 150 words per minute. This disparity is due, in part, to the limited number of brain activity sensors that can be applied to the brain’s surface. However, Cogan and his team aimed to overcome these limitations by collaborating with Jonathan Viventi, from Duke Institute for Brain Sciences, who specializes in developing high-density, ultra-thin, and flexible brain sensors.
The result of their collaboration is a remarkable implant consisting of 256 microscopic brain sensors, cleverly integrated into a postage stamp-sized, flexible, medical-grade plastic. These sensors enable the detection of distinct signals from neighboring brain cells, assisting in accurate predictions about intended speech.
To test the efficacy of their invention, Cogan and Viventi partnered with neurosurgeons from Duke University Hospital and temporarily implanted the device in patients undergoing brain surgery for unrelated conditions. During the experiment, participants were instructed to listen to a series of nonsense words and then repeat them aloud. The device recorded the activity from the speech motor cortex, responsible for coordinating the movement of nearly 100 muscles involved in speech production.
Remarkably, the speech decoder achieved an accuracy rate of 40%, despite operating with only 90 seconds of spoken data from a 15-minute test. Comparable brain-to-speech technologies typically require hours or even days of data to achieve similar results. Excitingly, the researchers are now working on developing wireless recording devices, eliminating the need for patients to be tethered to electrical outlets and allowing for greater mobility.
