Sign In

Communications of the ACM

ACM TechNews

Carnegie Mellon Researcher Creates Song-O-Matic

View as: Print Mobile App Share:

Carnegie Mellon University's Burr Settles has developed Titular, software that helps songwriters come up with titles for their new songs. Titular produces five new potential song titles each time the user clicks the "get new titles" button. Settles also developed LyriCloud, a program that provides new words to use in the lyrics of a song.

Settles created the songwriting tools using a linguistic model based on more than 137,000 songs from popular and obscure artists, and then tagging each word in the lyrics based on a part of speech.

"Writing lyrics that succinctly tell a story, express emotion, or create an image in the listener's mind is challenging," Settles says. "Even professional songwriters have trouble coming up with evocative titles or juicy words for the refrain. It seemed to me that the kinds of computational techniques we use to analyze natural language could be adapted to help overcome writer's block."

From Network World
View Full Article


Abstracts Copyright © 2011 Information Inc., Bethesda, Maryland, USA


No entries found

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