Artificial intelligence, really
Warner Robins teen engineers device to translate hand jive into spoken words
WARNER ROBINS --
Colby Wilkason is a real hands-on kind of young scientist.
The 17-year-old junior at Warner Robins High has parlayed her interests in science and music into a computer application that holds promise for deaf and blind people.
And it all started more than a dozen years ago when a precocious 4-year-old couldn't keep her hands to herself.
"She was a very busy kid and had already torn the wallpaper off in her room a few times," said Laura Wilkason, Colby's mother. "So we started her on piano lessons, just to give her something to do with her hands."
Upon further reflection some time later, Wilkason said, she and husband Tom came to realize Colby knew there was something underneath the wallpaper, "and she had to know what it was. Since then she has been taking things apart and then putting them back together. It's a good skill to develop."
Over time, Colby's musical and engineering skills meshed into her current science project, "Applications of Real Time Neural Networks for Use with Data Gloves." In everyday language, it's about teaching a computer to recognize - and speak - letters and words using American sign language.
Colby said she's been working on the project for about 2 years now, and plans on tweaking the project more for next year's round of regional, state and international science fairs.
"It's not done, even now, and I know I'll still be working on it in college," she said. "The technology found is not usually found, and it's really expensive."
The current setup grew out of a project she did for music notation, the teenager said, and evolved from notes to letters to words.
"It's a culmination of ideas and music therapy," Colby said. "I can't play the trumpet but I can make the computer make trumpet sounds," similar to a music synthesizer.
The project involves a significant amount of research and that can be time-consuming, especially with verifying the data. And she has tapped into the minds of scholars and professionals from Stanford University to the University of Sweden.
"Some of the material for the neural network came from Sweden. These are published works you can download from the Internet," she explained. "There really isn't that much information available, but I've been in contact with some people in artificial intelligence."
Colby adapted the neural network information to produce her own, and said she can "teach it anything" as far as sounds she wants to produce.
She donned the gloves and within a few seconds her hand motions elicited an audible "Hi" from the computer. Another flurry of motions and the computer intoned, "Go Demons." After all, this a serious Warner Robins High student.
Colby will take her award-winning project to the Intel International Science and Engineering Fair in Atlanta on May 11-16 to compete against 1,500 other young scientific minds.
Her project, already involving thousands of hours of work over the last 2 years, has been awarded a bevy of regional, statewide and national science and engineering fair honors - even though it's not yet finished.
And what has the young student got out of all this work?
"It's really opened my eyes to technology use and the ways it can be applied," Colby said. "Specifically, I'm in love with artificial intelligence because it helped me understand what science can do for us. It's a part of you."
When not consumed by research and experimentation with her project, Colby said she likes being with her friends, reading and playing tennis. She also plays oboe with the Mercer University and Macon Youth symphonies.