acm-header
Sign In

Communications of the ACM

ACM TechNews

AI Created More Than 100,000 Pieces of Music After Analyzing Irish and English Folk Tunes


View as: Print Mobile App Share:
Musical notation.

An artificial intelligence system developed at the KTH Royal Institute of Technology in Sweden generated more than 100,000 tunes after analyzing Irish and English music.

Credit: lilypad.org

A machine learning artificial intelligence (AI) system supervised by a researcher at the KTH Royal Institute of Technology in Sweden has generated more than 100,000 folk tunes after analyzing Irish and English music.

KTH's Bob Sturm said the project aimed to train computer models on folk music so they appear to have some musical intelligence, and then "devise methods to unravel what they are actually doing."

The AI employs a recurrent neural network to predict what comes next based on observation of a website with transcribed tunes.

Said Sturm, "The resulting computer models show some ability to repeat and vary patterns in ways that are characteristic of this kind of music. It was not programmed to do this using rules—it learned to do so because these patterns exist in the data we fed it."

The research yielded an album in which 50% of the tunes are computer-generated.

From Karlsruhe Institute of Technology
View Full Article

 

Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account