'New AI System Translates Brain Activity into Text, Opening Communication Possibilities'

'New AI System Translates Brain Activity into Text, Opening Communication Possibilities'
Published 1 years ago on May 24, 2023

The system shows promise in helping individuals who are mentally conscious but unable to physically speak, such as those affected by strokes, to communicate intelligibly once again.

Published in the journal Nature Neuroscience, the study utilizes a transformer model similar to those employed by OpenAI's ChatGPT and Google's Bard. The decoder measures brain activity using a functional MRI scanner after extensive training. Participants listen to hours of podcasts while inside the scanner, and later, the machine generates corresponding text based solely on the brain activity captured during the session.

Although the system does not provide a word-for-word transcript, it effectively captures the essence of what is being said or thought. After training the decoder to monitor participants' brain activity, it accurately reproduces the intended meanings of the original words around 50% of the time.

For instance, when a participant heard someone mention not having a driver's license, the system translated their thoughts as, "She has not even started to learn to drive yet." The researchers are working on refining the model to decode continuous language over extended periods, encompassing complex ideas.

Beyond stories, the researchers expanded their experiments to include silent videos. The semantic decoder successfully used participants' brain activity to describe specific events from the videos.

While the system's current reliance on fMRI machines limits its practical use outside the laboratory, researchers believe it could eventually be adapted for more portable brain-imaging systems. Addressing concerns about privacy and potential misuse, the team emphasizes the importance of proactive policies and regulations to protect individuals and ensure that these technologies are used only with their consent and benefit.

Study leader Jerry Tang, a doctoral student in computer science, stated, "We want to make sure people only use these types of technologies when they want to and that it helps them." The researchers also emphasized that the system cannot decode thoughts without the individual's knowledge and that certain techniques, such as thinking of animals, can protect against decoding attempts.

This pioneering research opens up exciting possibilities for improving communication and understanding for individuals with speech-related disabilities, and it lays the foundation for future advancements in brain-computer interfaces.


  • Written news comments are in no way https://www.showbizglow.com it does not reflect the opinions and thoughts of. Comments are binding on the person who wrote them.