Ideas, Tech, News and Fun

It may not be perfect but still revolutionary...

A new gadget is capable of brain-to-speech conversion. It uses brain signals to analyse what a person has thought to say out loud. Using this analysis it is able to emulate audible sentences what the person intended to say. This is a completely new component when it comes to brain to computer interface.

While the device is still quite buggy, as research is still in progress. In fact, the device can be thought off as an early prototype, however, it does its job. It becomes the first gadget to perform this novel function.

Best use

The device is a sign of hope for all the disabled who have lost their ability to communicate in sentences. It can also help people who stammer a lot or who have completely lost the ability to form sentences. Moreover, even people who could never speak at all may find hope that in future a more perfect device can help ease their problems.

How does it work?

While it may be a surprise but the device does not exactly use the information provided by the brain signals. THe brain signals are way too complex and may be for different instructions. Instead, the suystem uses Artificial Intelligence to combine the analysis of brain signals as well as the preliminary action of the vocal cords in trying to formulate the words.

In other words, the device is only able to predict and produce audible sentences as long as they are already about to be formed in our voice box. Currently, it is useful for only those people who, on some level, are able to form the words in their throats but can;t deliver them through their mouth.

When it was being tested, the device was correctly able to predict what people were speaking when they were asked to only mouth the words. Thus, the gadget is still working on a very basic level. However, it is expected to progress further.

Far from perfect

When the people testing the device were asked for the machine's accuracy, they responded that the sentences generated often got one word wrong more than half the times. A neuroengineer from Northwestern University, Marc Slutzky told SciAm:

For someone who’s locked in and can’t communicate at all, a few minor errors would be acceptable. Obviously you’d want to [be able to] say any word you’d want to, but it would still be a lot better than having to type out words one letter at a time, which is the [current] state of the art.