Researchers at the Institute for Mental BehaviorMortimer B. Zuckerman at Columbia University in New York announced the invention of the brain-computer interface, which translates human thought into speech with greater clarity and accuracy than existing solutions.
In a group report released TuesdayScientific Reports magazine tells how they created a system that can read the brain activity of a patient that is listening, and then repeat what this patient heard, with clarity never before seen in this technology.
Breakthrough opens the way for neuroprostheticsspeech or implants to have a direct connection with the brain. Ideally, technology will someday allow those who have lost the ability to speak to recover their voices. This can help patients who have had a stroke or are living with amyotrophic lateral sclerosis, easier to communicate with relatives.
Right now, technology can only sound what they are listening to, not what they want to say. But further development can improve the system for this purpose.
The technology is based on decades of research,which show that when people talk or listen or even imagine that they say or listen in the brain, certain patterns of activity arise.
Initially, researchers asked patientspeople suffering from epilepsy, who have already undergone surgery on the brain, to listen to the suggestions of various people, at this time, patterns of brain activity were measured. After the system was familiar with brain activity, patients were asked to listen to the recording of someone who pronounced numbers from zero to nine. Their brain activity was recorded and passed through the vocoder. The vocoder repetition was analyzed and cleaned using an artificial intelligence system, and as a result, a robotic voice sounded, repeating numbers from zero to nine.
Recording sounds was understandable to 75% of people, this performance is much higher than previous attempts.