Sign In

Communications of the ACM

ACM TechNews

What Will it Take to Make A.I. Sound More Human?

View as: Print Mobile App Share:
Pepper the robot with a Softbank executive at CES 2016 in Las Vegas.

Carnegie Mellon University professor Alan Black says conversation fillers are crucial to improving communications between humans and artificial intelligence.

Credit: James Niccolai

Conversation fillers such as "hmm" and "uh-huh" are critical to improving communication between humans and artificial intelligence (AI), according to Carnegie Mellon University professor Alan Black, who specializes in speech synthesis and ways to make artificially intelligence speech sound more real.

Although systems such as Siri and Cortana incorporate some of Black's work, the technologies simply respond to human speech. Black says the key to making AI conversations more natural are the pauses, fillers, laughs, and anticipation that help build rapport and trust. "You need mm-hmm, back channels, hesitations, and fillers, and so far our speech synthesizers can't do that," he says.

Black is experimenting with using voices recorded in dialog by modeling and incorporating the variance in human responses rather than using the same response all the time. Ultimately, he says AI systems could know the user's views on certain topics so it will not say something offensive.

"It's all about building this thing that's close to what humans expect and makes it easier to have this conversation," Black says.

From IDG News Service
View Full Article


Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account