
News
Share
Published 17:53 15 Sept 2022 BST
Updated 17:57 15 Sept 2022 BST

Erica in action (Image: Graduate School of Informatics, Kyoto University, Japan)[/caption]
In a report published in the Frontiers in Robots and AI journal, scientists explained how they have been trying to make Erica's conversations more natural.
Dr Koji Inoue, lead author and assistant professor at Kyoto University in the Department of Intelligence Science and Technology, said the team thinks one of the important functions of conversational AI is "empathy".
The author explained how: "Conversation is, of course, multimodal, not just responding correctly. So we decided that one way a robot can empathise with users is to share their laughter, which you cannot do with a text-based chatbot."
[caption id="attachment_359094" align="alignnone" width="498"]
The research focused on shared laughter (Image: Graduate School of Informatics, Kyoto University, Japan)[/caption]
Researchers focused on "shared laughter", where a user laughs and the system laughs in turn. They used AI to help the robot detect not just laughter, but whether it is appropriate to laugh, and what kind of laughter is best to use in the situation.
Erica took part in four, two-to-three minute dialogues with people which were then used to test the system. These dialogues went well, scientists said, but improvements need to be made to make the laughter truly natural.
Dr Inoue said: "Robots should actually have a distinct character, and we think that they can show this through their conversational behaviours, such as laughing, eye gaze, gestures and speaking style.
"We do not think this is an easy problem at all, and it may well take more than 10 to 20 years before we can finally have a casual chat with a robot like we would with a friend."
Related links: