Rhythmic similarity presentation

Annotated bibliography
  1. Foote, J., and S. Uchihashi. 2001. The beat spectrum: A new approach to rhythm analysis. Proceedings of the 5th International Conference on Multimedia and Expo.
    This paper describes a novel approach to evaluating rhythmic similarity by beat spectrum. The beat spectrum is a feature set based on the periodicity of FFT similarity.
  2. Foote, J., M. Cooper, and U. Nam. 2002. Audio retrieval by rhythmic similarity. Proceedings of the 3rd International Symposium on Musical Information Retrieval.
    This paper presents an evaluation of audio retrieval using the beat spectrum method. An accuracy of 96.7% was achieved when matching 15 patterns among the 4 originating songs.
  3. Paulus, J., and A. Klapuri. 2002. Measuring the similarity of rhythmic patterns. Proceedings of the 3rd International Symposium on Musical Information Retrieval.
    This paper describes an approach to evaluating rhythmic similarity by loudness, brightness, and MFCC feature extraction. Several proposed sub-components of the system (pattern segmentation, stochastic extraction) were not tested due to technical difficulties. Brightness, measured as spectral centroid, was found to be the most accurate feature set.