This is Erica: the robot capable of laughing thanks to Artificial Intelligence


Today we bring you the story of Erica. An automaton that seeks to be friendly through a believable meticulous laugh fueled by an artificial intelligence.

The future of humanity today, whether we like it or not, contemplates the integration of Artificial Intelligence (AI) and automated robots using said platforms.

There is evident concern about the direction these two factors may take and more than one member of the scientific community has expressed concern about the future.

Creating self-aware artificial intelligence and much more powerful and potentially lethal humanoid robots can unsettle more than one sensible human being.

So there is a group of scientists concerned with actively developing a solution to eliminate the barriers between robots and humans with one basic factor: laughter.

Erica laughs like a human so you don’t get upset

The official blog of the magazine Frontiers in Robotics and AI has just published the details of a recent research project, promoted by Japanese scientists from Kyoto University.

Where, basically, the engineers have developed a robot called Erica, which is powered by an Artificial Intelligence platform to which they try to teach the different nuances of laughter, from a very subtle one to a very loud one.

Frontiers in Robotics and AI

All with the aim of developing an efficient algorithm that helps improve natural conversations between people and artificial intelligence systems so that the experience is not disturbing.

The British newspaper Guardian It also details some details of this project. Where the robot Erica was programmed from data collected from 80 speed-dating dialogues between her and some college students to better understand the harmony of laughter in a talk:

“One of the important functions of conversational AI is empathy. So we decided that one way a robot can empathize with users is to share their laughter.”

These are the words of Koji Inoue, a doctor at Kyoto University and the main author of the research, where the data they collected during these dialogue sessions served to understand the rhythm and evolution where the right moment for solo laughter arises, social laughter and gleeful laughter in conversation.

Erica’s machine learning system is now able to decide whether or not to laugh and choose the appropriate type of sound based on the tone and pace of the conversation.

Leave A Reply