Brain Waves to Speech
A few weeks ago, I wrote about a device that allows the paralyzed to communicate through sniffing. While this invention is a major breakthrough for people with locked-in syndrome—such as sufferers of Lou Gehrig's disease—the task of sniffing out individual letters is tedious and time-consuming. Now scientists are working on an even better way for paralyzed people to communicate. A team of researchers at the University of Utah are beginning to decode speech directly from brain signals in an advancement that could help thousands of people.
The scientists were able to translate brain signals into words by implanting two grids of 16 microelectrodes between the skull and the brain. The results of the groundbreaking study are published in the September issue of the Journal of Neural Engineering."We have been able to decode spoken words using only signals from the brain with a device that has promise for long-term use in paralyzed patients who cannot now speak," says Bradley Greger, an assistant professor of bioengineering, in a University of Utah press release.
In the study, researchers placed microelectrodes over the speech centers in the brain of a volunteer with severe epilepsy. For his seizures, the patient had already undergone a craniotomy—surgery in which part of the skull is temporarily removed—so placing new electrodes into his brain was relatively easy.
With the microelectrodes, the scientists recorded the patient's brainwaves as he read 10 useful words: yes, no, hot, cold, hungry, thirsty, hello, goodbye, more and less. They then tried to discover which brain signals corresponded with which words. When the researchers compared any two words—such as "hello" and "goodbye"—they were able to distinguish corresponding brain signals 76 percent to 90 percent of the time. But when they examined all 10 words at once, they were only able to find the corresponding brain signal 28 percent to 48 percent of the time. Although this is better than random chance (which would be one in 10, or 10 percent of the time), the statistic is not quite good enough to create a translation device for paralyzed patients' brain waves.
"This is proof of concept," Greger says. "We've proven these signals can tell you what the person is saying well above chance. But we need to be able to do more words with more accuracy before it is something a patient really might find useful."
The scientists envision a wireless device that translates thoughts into computer-spoken speech. Such a machine would be life-changing for people paralyzed by stroke, Lou Gehrig's disease and other similar traumas. "Locked-in" patients today can only communicate through tiny movements—blinking an eye or moving a hand slightly.
Other members of the University of Utah team included electrical engineers Spencer Kellis, a doctoral student; Richard Brown, dean of the College of Engineering; and Paul House, an assistant professor of neurosurgery. Another coauthor was Kai Miller, a neuroscientist at the University of Washington in Seattle.
The research was funded by the National Institutes of Health, the Defense Advanced Research Projects Agency, the University of Utah Research Foundation and the National Science Foundation.
The scientists were able to translate brain signals into words by implanting two grids of 16 microelectrodes between the skull and the brain. The results of the groundbreaking study are published in the September issue of the Journal of Neural Engineering."We have been able to decode spoken words using only signals from the brain with a device that has promise for long-term use in paralyzed patients who cannot now speak," says Bradley Greger, an assistant professor of bioengineering, in a University of Utah press release.
In the study, researchers placed microelectrodes over the speech centers in the brain of a volunteer with severe epilepsy. For his seizures, the patient had already undergone a craniotomy—surgery in which part of the skull is temporarily removed—so placing new electrodes into his brain was relatively easy.
With the microelectrodes, the scientists recorded the patient's brainwaves as he read 10 useful words: yes, no, hot, cold, hungry, thirsty, hello, goodbye, more and less. They then tried to discover which brain signals corresponded with which words. When the researchers compared any two words—such as "hello" and "goodbye"—they were able to distinguish corresponding brain signals 76 percent to 90 percent of the time. But when they examined all 10 words at once, they were only able to find the corresponding brain signal 28 percent to 48 percent of the time. Although this is better than random chance (which would be one in 10, or 10 percent of the time), the statistic is not quite good enough to create a translation device for paralyzed patients' brain waves.
"This is proof of concept," Greger says. "We've proven these signals can tell you what the person is saying well above chance. But we need to be able to do more words with more accuracy before it is something a patient really might find useful."
The scientists envision a wireless device that translates thoughts into computer-spoken speech. Such a machine would be life-changing for people paralyzed by stroke, Lou Gehrig's disease and other similar traumas. "Locked-in" patients today can only communicate through tiny movements—blinking an eye or moving a hand slightly.
Other members of the University of Utah team included electrical engineers Spencer Kellis, a doctoral student; Richard Brown, dean of the College of Engineering; and Paul House, an assistant professor of neurosurgery. Another coauthor was Kai Miller, a neuroscientist at the University of Washington in Seattle.
The research was funded by the National Institutes of Health, the Defense Advanced Research Projects Agency, the University of Utah Research Foundation and the National Science Foundation.
Comments