Universal Whistling Machine, by Canadian artists Marc Böhlen and J.T. Rinker, is an attempt at developing a communication system that computers can understand, a tone-based interpreter of whistles.

Using a signal-processing computation system similar to the chips in mobile phones, the U.W.M. can extract whistles from other sounds, and exchange passages with humans, each other, and animals. Over time, it builds a database of every whistle it’s ever heard, increasing its vocabulary and range. What looks at first like a simple process becomes ever more interesting, a technical mocking bird that’s either mimicking or earnestly trying to communicate.

Video 1 and 2.

The work shared first prize at the Art & Artificial Life International Competition.