The shared-laughter AI system is being used to train a robot called Erica to detect laughter from those around 'her', then decide whether to laugh and what kind of laughter would be best, differentiating between light chuckles and rip-roaring peals of laughter.
The scientists' findings - described in the journal Frontiers in Robotics and AI - could in future help make conversations between humans and robots more natural.
“We think that one of the important functions of conversational AI is empathy,” said lead author Dr Koji Inoue. “Conversation is, of course, multimodal, not just responding correctly. 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.”
In the shared-laughter model, a human initially laughs and the AI system responds with laughter as an empathetic response. This approach required designing three subsystems – one to detect laughter, a second to...