A severely handicapped person communicates using brain signals sent to his or her vocal cords

A severely handicapped man has spoken with a type of technology that translates signals from his brain to his speech directly in words that appear on a screen. Produced by researchers in UC San Francisco, this method is the most natural way to communicate with people who are less able to communicate than other methods see you so far.

Meanwhile, neuroprosthetic technology has allowed organ users to write one letter at a time, a process that can be delayed and tedious. It also included other parts of the brain that control the arm or hand, a process that is not really good at this point.

The USCF system, however, uses a plug that is placed directly on the side of the brain dedicated to speech. In this way, the head can stimulate the brain to express itself, and the machine can translate the whole word, not just one letter, into a window.

In order to be effective, well-meaning patients have volunteered to re-evaluate their brain art. The researchers were able to analyze those species and devise new methods of identifying them in real time, using linguistic variables for accuracy.

However, the team was not sure if the neurotransmitters would remain strong in patients who had been paralyzed for years. To that end, he enlisted anonymous (Bravo1) collaborators who worked with researchers to create a 50-word word that the group could use on a computer. This included words like “water,” “family” and “good,” enough to allow the patient to make hundreds of decisions for use in their daily lives. The team also used “only accurate” terms similar to those found in voice recognition software.

To test the system, the team asked patient Bravo1 to answer questions such as “How are you today?” and “Would you like water?” The patient’s test was then displayed on the screen as “I’m fine,” and “No, I’m not thirsty.”

The machines were able to recognize their speech up to 18 words per minute with 93% accuracy, with an average accuracy of 75 percent. This may not sound as good as 200 words per minute which is possible with good speech, but it is better than the speed at which it appears previous neuroprosthetic systems.

“In our experience, this is the first demonstration of complete expression from the brain activity of a paralyzed person who is unable to speak,” said Edward Chang, MD, Chair of Neurological Surgery at UCSF and lead author of the study. “It shows that I promise to restore communication using a natural brain-talking machine.”

The team claims that the experiment represents the greatest evidence of this new type of “speech neuroprosthesis.” Afterwards, they plan to expand the test to have more students, while also working to increase word volume and improve pronunciation.