A laughing robot, maybe is seen as ominous, but would he look less creepy if he laughed at the right moment during the conversation? This is a theory that has recently been tested by scientists. A group of researchers at Kyoto University in Japan marveled at a laughing robot named Erika powered by a conversation-oriented artificial intelligence system.
Since laughter is a normal part of human dialogue, they reasoned, it might be useful to see how people respond to chatty robots they can also laugh with. Their results were published last week in the journal Frontiers in Robotics and AI.
Artificial intelligence is good at logic, but laughing? Not really. For starters, researchers have recognized that people laugh for different reasons, and that complicates the situation. To facilitate the work of the artificial intelligence system, they divided laughter into two categories: general social laughter, when the AI laughs in response to human laughter, and cheerful solo laughter, when the robot laughs in response to an object or laughs during a conversation.
Researchers taught an artificial intelligence model how and when to laugh by allowing it to participate in speed dating with university students. Erica was remotely controlled by an actress who spoke into a microphone and controlled physical movements such as head nods and other gestures.
Chats lasted between 10 and 15 minutes, and data were taken from 82 conversations. The researchers recorded the conversations using microphones and cameras and annotated them based on when people laughed in social media and solo, and how those laughs differed. This data was then used to train an artificial intelligence system to teach it when to laugh and what type of laugh to use. They then applied their collaborative laughter algorithm to existing conversational software and asked 130 volunteers to listen and rate how well the robot mimicked empathy, understanding, and human-likeness.
Overall, the researchers noted that in situations where shared laughter was appropriate, Erica and her algorithm did a good job of convincing people that they were paying attention to what was being said. But it had some drawbacks and limitations. Erika was good at responding to laughter, but she didn’t really know when to laugh on her own. The researchers wrote in their discussion that this may be because learning to respond to a prompt is easier than understanding why the content of a conversation might be funny.
Whether or not Erika has a sense of human humor is only part of a larger project that roboticists and engineers are taking on: giving robots social skills. Since 2017scientists worked on how to make a robot to laugh persuasive (major technology companies like Microsoft, IBMand Meta also interested in it). A month ago, Italian engineers debuted a robot bartender capable of conducting a social conversation (unfortunately this is delayed for the foreseeable future due to privacy concerns). The idea is to give robots facial expressions, body language, speech and the ability to to understand and responding to people’s demands will make them more attractive and better daily interaction.
But ultimately there may be a slippery slope away from the social exchange that seems natural to humans an obscure valley– script. There are also ethical issues with robots believable. Still, there are practical reasons to continue working in this area: Making talking robots less scary and more approachable by giving them the right human characteristics, experts say, will be particularly useful one day to integrate them into the health care system. hospitalityor something else maintenance-oriented industries.