Technology is revolutionizing our present and future phenomenally. It has brought a great comfort to our lives by providing us products like smartphones, drones, and 3D-printers. In a further extension of these tech inventions, scientists have now claimed that they have developed a device that decodes brain signals into sentences such as "Critical equipment needs special maintenance" and "Is this seesaw safe?" etc.
Which scientists were involved in developing this decoder?
As reported by the Washington Post and Nature magazine, a team of scientists led by neurosurgeon Edward Chang at the University of California, San Francisco, recorded the brain signals of five epilepsy patients who were undergoing brain surgery.
When Chang fed these recordings into a computer model developed with a human-like vocal system, it generated a synthesized speech which was half intelligible. The interesting thing about the experiment was that only those brain signals were picked that order vocal organs to move. In other words, scientists neglected abstract thoughts of the patients.
How did the scientists record brain signals?
In this experiment, the brain signals were recorded using ECoG (Electrocorticography). It is a flexible pad of electrodes that rests on the brain's surface. Previously, it has been used by scientists to record signals to control robotic arms. But, for the first time, ECoG was used for recording brain signals that stimulate speech.
What is the accuracy of this device?
To test the device's ability to decode signals into words, researchers played synthesized results in front of people selected from the crowd-sourcing website, Mechanical Turk. They were given the task to transcribe the speech using a pool of possible words.
Technology review points out that, "On average, people managed to understand about 50 to 70% of the words."
What are the views of experts about the decoder?
Andrew Schwartz, a tech researcher at the University of Pittsburgh sees this as groundbreaking stuff in brain-computer interface. He says, "Right now, this is probably the best work done in BCI (Brain-computer-Interface).
The accuracy could have been far greater if the scientists probed within the brain tissue and not overlying it."
This experiment is a huge step towards developing a system to help patients who are in a severe paralyzed state such as in comma. However, this aim still needs extravagant efforts as we haven't yet acquired the advancement required for brain-computer interface to accomplish this task.
"There is still a long way to go before this kind of synthesized speech is completely intelligible," says Mark Slutzky who is a neurologist at Northwestern University, while giving an interview to Nature magazine.
Nonetheless, adding to the device's worth, in future, this decoder could also be used in our daily gadgets to fulfill our ultimate goal in tech which is to send texts directly using brain signals.
