Delay and dictionary

A year after Stanford’s work, in 2024, Stavisky’s team Published Your own research on a brain-to-stay system, which competed to 97.5 percent of accuracy. “Almost every word was correct, but communicating on the lesson can be limited, right?” Staviski said. “Sometimes you want to use your voice. It allows you to create an interjection, it is likely to be less that other people disrupt you – you can sing, you can use words that are not in the dictionary.” But the most common approach to generate speech depends on synthesizing it from the text, which directly leads to another problem with the BCI system: very high delay.

In almost all BCI speech AIDS, sentences appeared on a screen after a significant delay, after the patient finished the words together in their brain. The speech synthesis part usually occurred after the text was ready, which further delayed. The brain-to-text solution was also suffering from a limited vocabulary. Such the latest system supported a dictionary of about 1,300 words. When you tried to speak a different language, use more detailed vocabulary, or even the unusual name of a cafe around the corner, the system failed.

Therefore, Vairagkar designed his prosthesis to translate the signs of the brain into sounds, not words – and do it in real time.

Downtrodden

The patient, who agreed to participate in Vairagkar’s study, was named T15 and was a 46 -year -old man suffering from ALS. David M. Brandman, a neurosurgeon and co-writer David M. Says Brandman, “He gets seriously paralyzed and when he tries to speak, it is very difficult to understand him. I have known him for many years, and when he speaks, I understand that maybe 5 percent of what he is saying, 5 percent of what he is saying.” Before working with the UC Davis team, T15 communicated using a gyroscopic head mouse to control the cursor on the computer screen.