With Technology, A Voice for Children with Disabilities
8.27.12 | Speech synthesis or text-to-speak (TTS) technology has been around for decades; however, there have always been inconsistencies with developing a natural-sounding synthetic voice – especially for young children.
New York Times reporter Emily Hager notes that most devices used by children rely on modified adult voices that end up sounding a lot like “adults on helium,” as David Niemeijer, chief executive for AssistiveWare, said.
That’s no longer the case, thanks to the app Proloquo2Go 2.1 that AssistiveWare developed with its partner, Acapela Group. The app features two voices recorded by a children—a boy and a girl’s —and is recommended for 6- to 14-year-olds. Proloquo2Go 2.1 costs $190 via iTunes; the new voices can be added at no charge if you already own Proloquo2Go.
Most companies do not offer children’s voices for their TTS software because of the difficulties that come with recording children, including the time it takes to record a “library of phases,” writes Hager. While Proloquo2Go 2.1 represents a big advancement in TTS technology, some parents are conflicted about the realism.
When Shanay Finney, 30, learned that her 10-year-old autistic son, Dahmier, might be able to have an age-appropriate voice, she reacted with mixed emotions. On one hand, the sound of his voice would be more normal, but on the other hand, all the other little boys using Proloquo2Go would have the same voice.
“I’m going to keep it real,” she said. “It’s not my son’s voice. But I know it might help.”
While sound engineering can manipulate the pitch of an adult’s voice to mirror a child’s, a base recording is still needed to recreate the natural sound of a child’s voice. To develop Proloquo2Go 2.1, audio engineers pieced together words from a large bank of several thousand phrases and terms, although it still requires hundreds of hours to record. Hager uses the example of synthetically creating the word “impressive” from the words “impossible,” “president,” and “detective.”
Proloquo2Go is used by tens of thousands of children with disabilities such as autism and cerebral palsy. According to Hager, the company estimates that the majority of its users are under the age of 18, which explains the need for children’s voices – especially considering that approximately 60 percent of Proloquo2Go’s demographic are under 11 years old.
The next step for companies that develop TTS software will be in predicting the user’s mood and tone ahead of time so as to elicit faster responses, and providing room for speech elements like sarcasm, which is virtually unattainable with TTS programs.
Leave a comment
Comments are moderated to ensure topic relevance and generally will be posted quickly.