A novel self-powered throat patch developed by researchers at the University of California has the potential to completely change communication for people lacking vocal cords. This innovative gadget uses machine learning to convert muscle motions into speech, providing a non-invasive substitute for conventional techniques. Jun Chen, an assistant professor of bioengineering at the University of California, Los Angeles, came up with the idea for the gadget in an attempt to relieve the strain that extended lectures were putting on his vocal cords.
Chen, alongside his colleagues, designed the patch to adhere to the throat, where it deciphers muscle movements into speech using AI technology. Remarkably lightweight at 7.2 grams, the patch is sweat-resistant and generates its own electricity through the wearer’s muscle movements, eliminating the need for a battery. This innovation was detailed in a study published in the Nature Scientific Journal.
The device comprises five thin layers, including responsive materials that detect subtle throat muscle movements. These movements produce electrical signals, which are then converted into speech by machine learning algorithms. The patch’s outer layers are made of flexible silicone, while the central layer contains silicon and micromagnets that create a magnetic field varying with muscle movements. The intermediate layers, composed of copper wire coils, transform these magnetic field changes into electrical impulses.
In their experiment, Chen and his team trained a machine learning algorithm using electrical impulses generated by the patch. Eight participants were asked to pronounce five short phrases 100 times each, allowing the algorithm to learn the specific muscle movements associated with each phrase. The results were impressive, with the algorithm achieving about 95% accuracy in translating electrical impulses into speech, both voiced and voiceless.
Despite these promising results, Chen acknowledges limitations in the study. It involved only eight participants, none of whom had speech disorders, and focused on a limited set of phrases. Additionally, the challenge of mass production remains. Nevertheless, this throat patch represents a significant advancement, especially considering that approximately 30% of people will experience voice disorders in their lifetime. Current alternatives like the electrolarynx are often invasive and costly, making this new device a potentially transformative solution for many.