The brain activity decoder can reveal the stories that people are thinking.
The semantic decoder, a new artificial intelligence system, can translate the brain activity of a person while they listen to a narrative or imagine telling one silently. This is then translated into a continuous flow of text. The researchers at The University of Texas at Austin developed a system that could help people who have strokes but are still mentally aware to speak again.
Jerry Tang, a computer science doctoral student, and Alex Huth – an assistant professor in neuroscience and computer sciences at UT Austin – led the study. The work is based in part on a model of a transformer, similar to those that are used by Open AI’s ChatGPT or Google’s Bard.
This system is non-invasive, unlike other decoding systems that are in development. Participants do not have to only use words on a list. After extensive training, the brain activity is measured by an fMRI scan. The individual must listen to podcasts for hours in the scanner. If the participant agrees to have their thoughts decoded later, the machine can generate text based on brain activity only if they listen to a story or imagine telling one.
Source:
https://medicalxpress.com/news/2023-05-brain-decoder-reveal-stories-people.html