In an exciting breakthrough, scientists have successfully created a noninvasive AI system capable of translating a person's brain activity into a stream of text.
Described as a "semantic decoder" and developed by researchers at the University of Texas at Austin, the system could eventually benefit patients who have lost their ability to communicate, according to a peer-reviewed study published on Monday in the journal Nature Neuroscience.
Once the AI-based system is trained, it can generate a stream of text when a participant is listening to or imagining telling a new story, CNBC reported.
To develop the system, the researchers used a transformer model similar to the ones that support OpenAI's ChatGPT,
The decoder was trained by having the study's participants listen to podcasts within a functional magnetic resonance imaging (fMRI) scanner, which measures brain activity.
Scientists hope to further refine the technology so that it can be more portable, ultimately becoming a solution for patients with communication challenges caused by paralysis, a degenerative disease, or similar conditions.
No comments:
Post a Comment