Papers‎ > ‎

A long-range self-similarity approach to segmenting DJ mixed music streams

Artificial Intelligence Applications and Innovations. Springer Berlin Heidelberg, 2013. 235-244.

Note that this paper has been superseded by the new version @ /papers/segmentationextended

In this paper we describe an unsupervised, deterministic algorithm for segmenting DJ-mixed Electronic Dance Music (EDM) streams (for example; podcasts, radio shows, live events) into their respective tracks. We attempt to reconstruct boundaries as close as possible to what a human domain expert would engender. The goal of DJ-mixing is to render track boundaries effectively invisible from the standpoint of human perception which makes the problem difficult.

We use Dynamic Programming (DP) to optimally segment a cost matrix derived from a similarity matrix. The similarity matrix is based on the cosines of a time series of kernel-transformed Fourier based features designed with this domain in mind. Our method is applied to EDM streams. Its formulation incorporates long-term self similarity as a first class concept combined with DP and it is qualitatively assessed on a large corpus of long streams that have been hand labelled by a domain expert.

Music Self-similarity

This paper is published in the Springer collection Artificial Intelligence Applications and Innovations 2013 --

  title={A Long-Range Self-similarity Approach to Segmenting DJ Mixed Music Streams},
  author={Scarfe, Tim and Koolen, Wouter M and Kalnishkan, Yuri},
  booktitle={Artificial Intelligence Applications and Innovations},
Tim Scarfe,
20 Jan 2014, 17:04
Tim Scarfe,
19 Sep 2013, 20:48
Tim Scarfe,
2 Oct 2013, 15:19
Tim Scarfe,
5 May 2013, 16:55