UCSF researchers have developed a brain machine interface for people with neurological conditions who have lost the ability to speak. This implant turns brain signals into recognizable synthesized speech by picking up on individual nerve commands and translating them to a virtual vocal tract. The system isn’t perfect just yet, as while it can accurately capture the distinctive sound of someone’s voice, the are times it produces garbled words. Read more for a video and additional information.
“Scientists are also testing denser electrodes on the brain interface as well as more sophisticated machine learning, both of which could improve the overall accuracy. This would ideally work with any person, even if they can’t train the system before using it in practice,” reports Engadget.