Can the brain computer interface directly translate our ideas into text Sohu (attached) – 2828创业网

Advanced | brain computer interface can be directly translated into the text of our thoughts? (of paper) – Sohu technology selected from Science Daily machine in the heart of Cao Rui, Wu Panni compiled participation: ever wondered if there is a device to decode your mind into the real speech or written words will be like? Perhaps this will enhance the performance of some existing equipment voice interface, may cause the change of language pathology, especially for those who have no language and exercise ability of "syndrome" is even more so in patients with atresia. "Christian Herff" in the author human neuroscience frontier commentary said: "so, I don’t need to ask" Siri, how’s the weather today "or" Google, "where can I go to have lunch, but to think I say these things can be". The idea of reading someone may still be in science fiction, but scientists have been able to decode the signals in the brain when we speak or listen. Herff and his co-author, Dr. Tanja Schultz, compared the advantages and disadvantages of using a variety of brain imaging techniques to capture and decode the neural signals in the brain. These techniques include functional magnetic resonance imaging (MRI), near infrared imaging (according to the metabolic activity of neurons to detect neural signals), electroencephalogram (EEG) and magnetoencephalography (MEG) (to correspond with the voice of the electromagnetic activity of neurons detected). In particular, a method called the cerebral cortex electrogram (ECoG) has shown great promise in the study of Herff. This study presents a brain to text (Brain-to-text) system, which has been implanted in an electrode grid to treat patients with epilepsy. They read the text on the screen in front of them, and their brain activity was recorded. This forms the basis of the database that matches the language phonemes (or "speech") to the neural signal patterns. When the researchers applied both the language and the dictionary model to the algorithm, it was possible to decode the neural signal with high accuracy. "This is the first time that we have been able to use automatic speech recognition (ASR) technology in the brain signal, which can accurately decode brain activity," Herff said. But now the demand for implantable electrodes can be seen, this technology will take a long time to be used in daily life. So, what would you like to do if you want to develop a normal thinking testing device from now on? Herff said: "the first milestone is to decode the imaginary phrases from the brain, but we still have a lot of technical problems to solve. "The results of this study are very exciting, but for this type of brain computer interface research is still in the preliminary stage. Abstract: speech interface has been widely accepted and used in n相关的主题文章: