Brain scans can translate a person\u2019s thoughts into words
\nA new system was able to capture exact words and phrases from the brain activity of someone listening to podcasts.<\/p>\n
A noninvasive brain-computer interface capable of converting a person\u2019s thoughts into words could one day help people who have lost the ability to speak as a result of injuries like strokes or conditions including ALS. <\/p>\n
In a new study, published in Nature Neuroscience today, a model trained on functional magnetic resonance imaging scans of three volunteers was able to predict whole sentences they were hearing with surprising accuracy\u2014just by looking at their brain activity. The findings demonstrate the need for future policies to protect our brain data, the team says.<\/p>\n
Speech has been decoded from brain activity before, but the process typically requires highly invasive electrode devices to be embedded within a person\u2019s brain. Other noninvasive systems have tended to be restricted to decoding single words or short phrases. <\/p>\n
This is the first time whole sentences have been produced from noninvasive brain recordings collected through fMRI, according to the interface\u2019s creators, a team of researchers from the University of Texas at Austin. While normal MRI takes pictures of the structure of the brain, functional MRI scans evaluate blood flow in the brain, depicting which parts are activated by certain activities.<\/p>\n
First, the team trained GPT-1, a large language model developed by OpenAI, on a data set of English sentences sourced from Reddit, 240 stories from The Moth Radio Hour, and transcriptions of the New York Times\u2019s Modern Love podcast. <\/p>\n