Speech synthesized from cerebral activity



[ad_1]

Please consider that scientists create speech signals from the brain.

Scientists report having developed a virtual prosthetic voice, a system that decodes the vocal intentions of the brain and translates them into understandable words, without the need to move a muscle, even those of the mouth. (Physicist and author Stephen Hawking used a cheek muscle to type keyboard characters, synthesized by a computer.)

"It's great work, and it helps us restore the word," said Dr. Anthony Ritaccio, neurologist and neuroscientist at the Mayo Clinic in Jacksonville, Fla., Who was not a member of the research group. .

The new system, described Wednesday in the journal Nature, decrypts the brain motor drives guiding vocal movement during the speech – the lick, the narrowing of the lips – and generates intelligible sentences that are close to the natural cadence of the speaker.

Previous implant-based communication systems produced about eight words per minute. The new program generates about 150 minutes, the pace of natural speech.

The researchers also found that a synthesized vocal system based on a person's brain activity could be used and adapted by another person, indicating that ready-to-use virtual systems might be available someday.

The team plans to move to clinical trials to further test the system. The biggest clinical challenge may be finding the right patients: Strokes that inhibit a person's speech often damage or destroy areas of the brain that promote speech articulation.

It's a fascinating search. Congratulations to the researchers.

Mike "Mish" Shedlock

[ad_2]

Source link