Stanford researchers have developed a brain-computer interface (BCI) that can translate imagined words directly from neural activity into speech. This breakthrough marks the first time scientists have understood what brain activity looks like when someone thinks about speaking.
The BCI system uses microscopic electrode arrays implanted in the motor cortex, which directs movements involved in speech. It records activity while participants are presented with tasks such as attempting to speak aloud and silently imagining specific words. Machine learning models then classify distinct patterns of brain activity linked to phonemes, the smallest sound units in spoken language.
The system can recombine these phonemes into whole words and sentences in real time, reaching accuracy rates of up to 74 percent. While attempting speech produced a weaker signal, it was still distinguishable from imagined speech.
This technology has significant implications for individuals with severe paralysis, such as those living with amyotrophic lateral sclerosis or brainstem stroke. It could make communicating “easier and more natural” and restore conversational communication.
The researchers also discovered an important privacy concern: the system sometimes detected words that participants had not been asked to think about. To address this, they created a form of mental lock, which remains inactive unless triggered by an imagined password.
This breakthrough comes amid growing interest in BCIs from both academic and commercial sectors. While still experimental, the Stanford work provides proof-of-principle for future devices that could let users speak fluently using thought alone, offering “real hope” that speech BCIs can restore communication as fluent and natural as conversational speech.
Source: https://www.techspot.com/news/109081-stanford-brain-computer-interface-turns-inner-speech-spoken.html