This Algorithm Is Good Enough to Fool the Human Ear

Researchers at MIT have created robots that pass an audio Turing Test, fooling humans into thinking that artificially-created sounds are natural. A robot's sounds are crucial to its interactions with people. The sounds a robot makes can identify its purpose and usefulness. But current "approaches in AI only focus on one of the five sense modalities, with vision researchers using images, speech researchers using audio, and so on," says Abhinav Gupta, an assistant professor of robotics at Carnegie Mellon University not involved in MIT's work. "This paper is a step in the right direction to mimic learning the way humans do, by integrating sound and sight."