Communications of the ACM
Sound, Digested
Credit: rosiemoulton.wordpress.com
Backed by funding from the U.S. National Science Foundation (NSF), Imagine Research has developed MediaMined, an artificial intelligence system for understanding and indexing sound.
MediaMined "allows computers to index, understand, and search sound--as a result, we have made millions of media files searchable," says Imagine Research CEO Jay LeBoeuf.
MediaMined enables music producers to quickly scan a large set of tracks and recordings and automatically label the inputs. The software detects the properties of the sound wave represented by an audio file's data. The second stage of processing includes applying statistical techniques to estimate how the characteristics of the sound file might relate to other sound files. Because the software's sound database continues to grow, the characterization ability of the software improves as the program attracts more users and analyzes additional files.
"There are many tools to analyze and index sound, but the novel, machine-learning approach of MediaMined was one reason we felt the technology could prove important," says NSF's Errol Arkilic. "The software enables users to go beyond finding unique objects, allowing similarity searches--free of the burden of keywords--that generate previously hidden connections and potentially present entirely new applications."
From U.S. National Science Foundation
View Full Article
No entries found