A novel artificial intelligence system, the semantic decoder, can translate brain activity into continuous text. The system could revolutionize communication for people unable to speak due to conditions like stroke.
A new AI system, called a semantic decoder, can translate brain activity into text without the need for surgical implants. Developedby researchers at The University of Texas at Austin, this system has the potential to help people who are mentally conscious but unable to physically speak, such as stroke patients. The decoder utilizes an fMRI scanner to measure brain activity after extensive training. By listening to stories or imagining telling them, participants can communicate through the machine-generated text. This breakthrough allows for continuous language decoding with complex ideas, surpassing previous methods that focused on single words orshort sentences. While not word-for-word, the generated text captures the intended meaning of the original words. The research team addressed concerns about misuse, ensuring the technology is used voluntarily and beneficially. Although currently limited to lab use due to reliance on fMRI, the system shows promise for future application on portable brain-imaging systems. The study was supported by various foundations and co-authored by Amanda LeBel and Shailee Jain. Alexander Huth and Jerry Tang have filed a related patent application.
