A research team from the University of Texas at Austin has developed an artificial intelligence (AI) based "brain reading technique" that can convert human thinking into continuous text without the need for language. This brain decoder only takes about an hour to adapt to a person's unique brain activity pattern, providing an unprecedented new tool for language disorder patients and laying the foundation for the development of future neural technologies and next-generation brain computer interfaces. The relevant research results were published in the latest issue of the journal Contemporary Biology. Traditional brain computer interface technology relies on long-term data collection and complex algorithm training to recognize and interpret brain signals. This method is not only time-consuming, but also requires extremely high tolerance from participants. The key to the new tool being considered a major breakthrough is that it adapts to everyone's brain activity patterns very quickly. In contrast, early techniques required participants to lie in a functional magnetic resonance imaging scanner for 16 hours to train brain decoders. The new technology has created a language based brain computer interface method without requiring any level of language understanding. The researchers used the method of watching silent short videos and combined it with a converter algorithm to map the brain activity of new individuals onto the dataset previously used to train brain decoders. This allows new users to use decoding technology with only one hour of training time. In addition, under simulated conditions of brain damage in aphasia patients, the improved decoder can convert the stories they are thinking in their minds into continuous text. This demonstrates the potential of the tool to assist patients with aphasia. Currently, researchers are collaborating with aphasia experts to test whether this improved brain decoder can help improve communication skills in aphasia patients. This study also means that accessing semantics through language and visual pathways has extraordinary value, and related technologies can open up new possibilities for neuroscience research. At the application level, it will soon help those who find it difficult to speak or understand language. (New Society)