When Speech is Not a ‘Voice’

From front to back: Heather Warren-Crow, assistant professor of art theory and practice; Yi Hu, assistant professor of electrical engineering and computer science; Shelley Lund, associate professor of communication sciences and disorders; and Patricia Mayes, associate professor of English

When Speech is Not a ‘Voice’

What does it mean when the voice that expresses your thoughts isn’t your own? A transdisciplinary team of UWM researchers are looking at how synthetic voices impact those who use them and those they communicate with.

If you want 21st-century solutions to 21st-century challenges, you need to bring together researchers from disciplines that haven’t traditionally worked together. That’s the thinking behind the Transdisciplinary Challenge Awards funded through UWM’s Center for 21st Century Studies.

This year’s award is supporting an engineer, a speech-language pathologist, an artist-scholar and a linguist who are collaborating to explore the social and technological impacts of using synthetic voices. People with communication disabilities use these voices to communicate through augmentative and alternative communication (AAC) technologies.

While voices in early devices sounded robotic, newer voices are more natural-sounding. However, AAC technologies still have challenges, like a limited number of voices to choose from, says Shelley Lund, associate professor of communication sciences and disorders.

Synthetic voices are now easier to understand, but other issues remain – such as how to convey emotion and personality, and whether natural-sounding voices are even preferable.

“Emotion is the most challenging aspect of the voice to convey through synthesis,” says Heather Warren-Crow, an assistant professor of art theory and practice.

“We want to look at the issues people using AAC technologies face – how the use of specific available voices affects their identities and their interaction with others,” adds Patricia Mayes, a linguist in the English Department.

The researchers will interview AAC users about their perceptions and preferences, and also record their interaction with others to determine how to improve synthesized speech.

Researchers will refine and perhaps expand the scope of a tablet application they are developing under the leadership of team member Yi Hu, an assistant professor of electrical engineering and computer science. “We want to see if we can convert what we learn into practical applications for devices like smart phones,” he says.

“Starting from the perspective of the humanities and arts, our goal is to improve the devices created by engineers and designers,” says Mayes.

Augmentative and alternative communication (AAC) technologies were first developed 40 years ago and are now available on tablet computers such as iPads. They allow users with communication disabilities to type what they want to say into a device that converts their message to spoken words.