Meta working on AI tech that can read brain waves to ‘hear’ what people can hear

In another futuristic endeavor, Facebook parent company Meta is currently developing artificial intelligence that can read human brainwaves to “hear” what a person is hearing in real time.

The early research stages included studying 169 adult participants, who were connected to electrodes and non-invasive devices to record the participants’ brain waves as they listened to stories read to them as well as sentences in English and Dutch.

Their responses were recorded and the information fed into an artificial intelligence model to sift through the information and filter out patterns to determine what the person was hearing.

The idea is to “decode speech from noninvasive recordings of brain activity,” which Meta hopes can, in the future, assist those affected by traumatic brain injuries that leave them unable to communicate via talking or typing. The early research looks promising, according to Meta.

“From three seconds of brain activity, our results show that our model can decode the corresponding speech segments with up to 73 percent top-10 accuracy from a vocabulary of 793 words, i.e., a large portion of the words we typically use on a day-to-day basis,” the company said in a statement.

Meta hopes the new tech could have impactful real-life applications in the future. “The ultimate goal of enabling patient communication will require extending this work to speech production. This line of research could even reach beyond assisting patients to potentially include enabling new ways of interacting with computers,” the company said.

What patients need down the line is a device that works at bedside and works for language production,” said Jean Remi King, a research scientist at Facebook Artificial Intelligence Research (FAIR) Lab to TIME.

“I think one possible next step is to try to decode what people attend to in terms of speech—to try to see whether they can track what different people are telling them,” she added, “but more importantly, ideally we would have the ability to decode what they want to communicate.”