Facebook is getting closer to making its brain-reading computer a reality.
Facebook Reality Labs, a division that studies long-term tech products and hardware, has been working with researchers from the University of California, San Francisco on the technology. A study published Tuesday in the journal Nature Communications showed how researchers were able to use the brain-computer interface to decode speech directly from the human brain onto a screen.
For the study, researchers worked with three patients undergoing treatment for epilepsy who had electrodes implanted into their brains. Researchers will spend the year testing the technology.
UCSF researchers say the findings could help develop a communication device for patients who can no longer speak after severe brain injuries, including brainstem stroke and spinal cord injury.
But the successful trials are more likely to be relevant to Facebook's efforts in developing augmented reality glasses.
"Today we're sharing an update on our work to build a non-invasive wearable device that lets people type just by imagining what they want to say," Andrew "Boz" Bosworth, Facebook vice president of AR/VR, said in a tweet. "Our progress shows real potential in how future inputs and interactions with AR glasses could one day look."
Facebook said it doesn't expect this technology to be available to consumers anytime soon.
"It's currently bulky, slow, and unreliable," the company said. "But the potential is significant, so we believe it's worthwhile to keep improving this state-of-the-art technology over time."
Facebook isn't the only Silicon Valley company interested in the promise of brain-computer interfaces.
Elon Musk's brain-computer start-up Neuralink has made headway on its efforts in the space. Earlier this month, the Tesla CEO said at a Neuralink event that the company intends to begin human trials before the end of 2020.