Arsh Shah Dilbagi, a teenage student from Panipat, India, has taken on an ambitious project to change the way people suffering from speech impairments communicate with the world.
Intended for users with developmental disabilities like Down's Syndrome, ALS or other challenges, Dilbagi says his device could even bring the ability to communicate aurally to mutes.
Dilbagi's TALK is an Augmented and Alternative Communication (ACC) device capable of analyzing the breath patterns from a user and translating them into Morse code which is then translated again to create the intended 'speech.'
A 16 year-old student at DAV Public School, he is a roboticist at heart who loves making things.
"I've always been fascinated by the power of Science and Mathematics," he says. "In 2010 my parents gifted me a LEGO Mindstorms Kit and since then I have never stopped making and learning."
In the past, the young scientist and inventor conceived and designed a working prototype of an autonomous UGV which earned him kudos from no less of an eminence than the President of India.
As for his latest project, he says he believes it will bring a whole new level of convenience to a community of people who are largely forgotten.
"TALK has the potential to change the world," Dilbagi said. "By enabling people with disorders like LIS, ALS etc., speech impairments like Dysarthria and even mutes to communicate and interact with the world like never before."
Inspired by inventors and entrepreneurs like Steve Jobs and scientists like Stephen Hawking, Dilbagi hopes to pursue robotics in the future, and he sees a win in the Google Science Fair contest as a stepping stone to that end.
His TALK device works like this: a sensor is placed under the nose and the user then creates a series of short and long exhales. That data is sent to the computer via USB port using an Arduino board and then logged in a text file. He says that as the data reflected a definitive pattern, he continued to develop the system.
To do that, Dilbagi chose a random sentence with 50 letters and then dictated it using the TALK device in a controlled environment with very little noise and again in "natural surroundings" which included ambient noise. He says that once he discovered that the average accuracy was very high, bordering on 99%, it was time to see if it could be produced for less than $100.
Dilbagi says his device differs from current ACC devices in how it decodes inputs from the user. The current technology works on tracking eye movements, while his device uses breath analysis through sensors attached to a microphone worn by the user.