More than a half-million children under age 15 have a severe communication disorder, impairing their ability to speak or communicate with others.
Now, advances in technology are giving them a voice — some, like Shannon Ward, for the first time.
When Shannon came into the world three months early, her mom knew she’d be special.
"She was always just aware, alert, and watching," said Janine Blythe, Shannon's mom.
However, while her mind was active, Shannon’s physical development was delayed. Diagnosed with cerebral palsy, Shannon couldn’t speak.
"She would vocalize for you, squeal with delight, but she could not say, mom or dad," Blythe said.
At age nine, she was using assistive technology to help her communicate. Now, she is a teenager.
"I think everyone deserves to be heard," Shannon said.
The adult robotic voice doesn't match Shannon's age or personality.
"There are not really a lot of choices for these kids," Blythe said.
That's why Tim Bunnell and his team at Nemours are working to give kids like Shannon a voice of their own for the first time.
"We take recordings of natural speech, chop them up into very small pieces, and then paste them back together again in novel ways," said Dr. Tim Bunnell, principal research scientist and director, Center for Pediatric Auditory and Speech Sciences, Nemours/Alfred I. duPont Hospital for Children.
For those who can't speak words, the focus is on vowel-like sounds.
"Perceptually, we are tuned to listen to the vowels to understand the vocal identity of a person," Bunnell said. "For Shannon, it was as sound like 'Ahh.'"
Characteristics of those sounds, such as the child's pitch and voice quality, are recorded and blended with a donor child's voice to build a new voice.
“I really like my new voice. I think everyone deserves to be heard and it feels great to have a voice that matches who I am," Shannon said.
The technology used to create a new voice for children like Shannon with cerebral palsy was originally developed for voice banking — so that those with Lou Gehrig's disease can save their voices for later use.
In the next three-to-five years, the technology will be able to actually calculate the child's voice as they age, so their voice will grow along with the child.