web analytics

Teenager Develops Device that Allows User to Communicate with Just Exhales

AAC device

Image credit: Components of Talk’s final design / Arsh Shah Dilbagi’s Google Science Fair project

In order to participate in the Google Science Fair, Arsh Shah Dilbagi, a 16-year-old teenager from Panipat, India, invented an apparatus which could enable patients of speech impairment or paralysis to communicate by means of only two different kinds of breaths.

At present, many augmentative and alternative communication (short-termed as AAC) devices available for the patients of speech impairments, locked-in syndrome, or ALS could be so expensive and huge in the customized manner that it could not applied to  everyone’s requirement. Therefore, Arsh Shah Dilbagi, known as budding roboticist to others, created a portable AAC device characteristic of being generic and rapid as well as affordable to ordinary users. He named his device as Talk, which consisted of two major parts: a processing unit similar to the size of a smartphone and a wearable sensor placed near the nose.

According to Arsh’s description of his project, Talk could turn two kinds of exhales, which would be distinguished in terms of their unique duration and intensity, into electrical signals by application of a microphone chip capable of sensing the pressure of the breaths.

Such electrical signals would be then processed by a microprocessor, which could mark the short exhales as “dots,” and the longer exhales as “dashes” respectively. Just like Morse code and binary languages, these two kinds of breaths could be translated into words and sentences. Afterwards, these words and sentences would be relayed to another microprocessor to be synthesized.

With two modes, Talk could communicate in English in one and give specific commands or phrases in another. Surprisingly, such device had nine different voices for males and females of different ages as well.

To test his device, Arsh had put it on himself, his family, and one patient of Parkinson. In his experiment, all involved were able to release two distinguishable signals by using their breath. It was amazed that the average accuracy of device was 99 percent. The cost of final version of the Talk would be less than $100. It was hoped by Arsh that he would do more by adding auto-predictions to Talk to integrate his device with smartphones and Google Glass.

Check out Arsh’s amazing video here!