acm-header
Sign In

Communications of the ACM

ACM TechNews

Enhanced 'Listening Skill' Makes Robot Better at Talking


View as: Print Mobile App Share:
A robot that listens better can better handle its part of a conversation.

Researchers at Osaka and Kyoto universities in Japan have given a conversational human-like robot greater ability to have a natural dialogue with people by enhancing its listening skills.

Credit: moneycontrol.com

Researchers at Osaka and Kyoto universities in Japan have upgraded a conversational human-like robot, giving it greater ability to have a natural dialogue with people by enhancing its listening skills.

They say the ERICA robot was specifically designed for natural conversation through the incorporation of human-like facial expressions and gestures.

In developing the upgrade, the researchers focused on creating a system for "attentive listening," meaning that a listener asks elaborating questions, or repeats the last word of the speaker's sentence, allowing for more engaging dialogue.

The team used a series of distance sensors, facial recognition cameras, and microphone arrays to collect data on parameters necessary for a fluid dialogue between ERICA and a human subject.

The robot's responses are generated through machine learning using counseling dialogue texts, resulting in dramatically improved dialogue engagement.

Said Kyoto University’s Tatsuya Kawahara, "Making a human-like conversational robot is a major challenge. This project reveals how much complexity there is in listening, which we might consider mundane."

From Money Control
View Full Article

 

Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA


 

No entries found

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