Beat Spectrum

For a number of years I worked on an approach to analyze audio and video based on self-similarity. Comparing every part of a media file exhaustively with every other part results in a "similarity matrix" that both visualizes the time structure of the media and facilitates further analysis like segmentation or verse/chorus detection in popular music.

A particular innovation is the "beat spectrum:" a measure of self-similarity of rhythmic audio versus lag time, which can describe the rhythmic character such as tempo, time signature, and "swing."

Along with my colleague Matthew Cooper, we produced a fruitful number of publications and patents exploring the applications of this approach.








More at

U.S. Patent 7,068,723: Method for automatically producing optimal summaries of linear media

U.S. Patent 7,284,004: Summarization of digital files

U.S. Patent 7,424,150: Systems and methods for media summarization