To communicate with humans, robots learn to laugh at the right time

To communicate with humans, robots learn to laugh at the right time

Anyone who has shared a friend’s laughter knows how deeply interconnected humor can be. So it makes sense to our future robot buddies You have a better chance of winning our trust and affection if they can laugh with us.

but only because a file The robot tells jokes Not that he can respond to it appropriately. Did the comment include a polite robot laugh or an all-out robotic laugh? A correct response could mean the difference between an accessible robot and a weak metallic one.

That’s why Japanese researchers are trying to teach nerdy robotics to laugh at the right time and in the right way. It turns out that training an AI to laugh isn’t as simple as teaching it to respond to a desperate phone tree call to unsubscribe. says The study was published Thursday In Frontiers in Robotics and AI.

Robot Erica looks straight ahead with a slight smile

Erica the humanoid robot in the lab has a good sense of humor.

Osaka University, ATR

The study details the team’s research into developing an AI conversation system that focuses on shared laughter to make chatter between humans and bots more natural. They envision it being integrated into existing chatbots for bots and agents, who are already learning Emotions revealed And the Dealing with open complexity Like mysterious human orders.

“We believe that one of the important functions of conversational AI is empathy,” Koji Inoue, assistant professor of informatics at Japan’s Kyoto University and co-author of the study, said in a statement. “Conversation, of course, is multimedia, it just doesn’t respond correctly. So we decided that one of the ways a bot can empathize with users is by sharing their laughter.”

The key is that the system not only recognizes laughter, but also decides whether to laugh in response and then chooses the appropriate type of laughter for the occasion. “The most important result of this paper is that we have shown how we can combine all three of these tasks into one robot,” said Inoue. “We believe that this kind of joint system is essential for appropriate laughter behavior, not just laughter detection and response.”

To collect training data on the frequency and types of shared laughter, the team bugged Erica, an advanced human-like robot designed by Japanese scientists Hiroshi Ishiguro and Kohei Ogawa, as a platform for studying human-robot interaction. Erica can understand natural language, has a complex human voice and can blink and move her eyes when listening to humans talking about their problems.

The researchers recorded a dialogue between students from Kyoto University, who took turns chatting face-to-face with Erica, as amateur actresses in another room hooked up the robot through a microphone. The scientists chose this setting knowing that it is normal for there to be differences between how humans talk to each other and how they talk to robots, even those controlled by another human.

“We wanted, as far as possible, to train the laughter model under conditions similar to real human-robot interaction,” Kyoto University researcher Divesh Lala, another co-author of the study, told me.

Left, a human talks to the robot Erica, who is controlled by an actress from a separate room.

Kyoto University

Based on the interactions, the researchers created four short audio dialogues between humans and Erica, which were programmed to respond to conversations with varying levels of laughter, from none at all to repeated chuckles in response to her fellow human conversational companions. The volunteers then rated these breaks on empathy, naturalness, similarity to humans, and comprehension.

The shared laughter scenarios performed better than those in which Erica neither laughs nor ever laughs every time she detects a human laugh without using the other two subsystems for context and response filtering.

Kyoto University researchers have already programmed their shared laughter system into robots alongside Erica, though they say a human’s howl could be a more natural sound. In fact, even as robots become increasingly lifelike, Sometimes disturbinglyRobotics scientists acknowledge that cultivating their own distinct human traits poses challenges beyond coding.

“It may take more than 10-20 years before we can finally have a casual conversation with a bot like we do with a friend,” Inoue said.

Needless to say, Erica is not yet ready for the standing rink. But it’s interesting to think that there may soon come a day when you really feel like she’s getting your jokes.

#communicate #humans #robots #learn #laugh #time

Leave a Comment

Your email address will not be published. Required fields are marked *