This entire album was composed using artificial intelligence

The humans behind it hope to release the album by the end of the year.

Robots can mimic human emotions, paint masterpieces, fold your laundry and even pay your insurance claim. But can they redefine pop music?

If the folks at Flow Machines get their way, we'll all be jamming along to an entirely artificial-intelligence-based music album in the very near future. The project just released a single, called "Daddy's Car," which you can listen to in the video above. Don't know about you, but we're picturing ourselves on a boat on a river, with tangerine trees and marmalade skies ...

Flow Machines, which operates out of the Sony Computer Science Laboratories in Paris, is trying to create the first pop album composed entirely by artificial intelligence. (Yes, not even Daft Punk can lay claim to that.)

French composer Benoit Carré is at the helm with the assistance of Israeli jazz saxophonist Gilad Atzmon. The company is trying to use a complex algorithm to allow a machine to produce music on its own, from start to finish. All they have to do is press play.

Jazz musician Gilad Atzmon is a music consultant for Flow Machines.Gilad Atzmon is a London-based jazz artist who was born in Israel and trained at the Rubin Academy of Music in Jerusalem. (Photo: Courtesy)

Using Flow Machines' algorithm, the computer can recognize patterns from the thousands of songs its developers have fed into it. It then uses those patterns to make autonomous decisions. It's remarkably self-aware, too.

Musician Benoit Carré is the artistic director of Flow Machines.Musician Benoit Carré is the artistic director of Flow Machines. (Photo: Benoit Carré Facebook)

“We don’t give the machine musical rules or abstract musical knowledge,” senior researcher Pierre Roy told the Seeker news site. “It’s only the machine producing music based on what it learned from the data.”

MORE FROM THE GRAPEVINE:

Photos and SlideshowsPhotos and Slideshows

Related Topics: Music

comments powered by Disqus