May 28, 2019
0 0

Translated brain signals into words

Written by

Some scientists have worked on a decoder that in the future could allow those who cannot talk through brain impulses.

A group of researchers from the University of California, San Francisco , coordinated by Gopala Anumanchipalli, developed a decoder with artificial intelligence that manages to translate into words the brain activity. It could be used by all those people who cannot express themselves, such as those suffering from Parkinson’s disease or Amyotrophic Lateral Sclerosis (SLA), who in the course of time deprive them of these abilities.

The realization took place by recording the activity of the cerebral cortex of 5 volunteer patients. These during the recording spoke aloud, while the researchers analyzed the signals that regulate the movements of the organs of language. Thanks to a synthesizer and AI, it was possible to convert impulses into words and sounds. 101 phrases were articulated during the search and those who heard them were able to understand and transcribe them.

This research will lead in the coming years to the development of instruments that might be able to translate your thoughts into words. Real neuro-prostheses that will allow to express themselves to those who cannot, but also to control objects like robotic limbs. This is certainly not the only research in this sense: only a few months ago at Columbia University a particular technology was created that allows to translate thoughts into words “reading” in the mind of a person. In this case a vocoder was used, used to decode the speeches and synthesize them.

A probable future opens up in which one can communicate directly with thoughts, even if it is not as close as we imagine. Even Elon Musk himself spoke of his Neuralink , which basically wants to fuse the human brain with artificial intelligence, for the company that deals with creating a man-machine interface.

Article Tags:
Article Categories:

Leave a Comment

Your email address will not be published. Required fields are marked *