A Two-Layer Approach for Multi-Track Segmentation of Symbolic Music

B. Rafael and S.M. Oertl (Austria)


pattern recognition, music segmentation, machine learning, music information retrieval


Music segmentation is a key issue in music information retrieval (MIR). Structural information about a composition achieved by music segmentation can improve several tasks related to MIR such as searching and browsing large music collections, visualizing musical structure, lyric alignment, and music summarization. Various approaches have been introduced to MIR, most of them concentrating on digital audio. The authors of this paper present a two-layer approach for the segmentation of symbolic music. The first step uses exact and approximate string matching methods to detect the best candidate segmentations for each track of a multi-track composition independently. The second step combines all single track results and determines the best segmentation for each track in respect to the global structure of the composition.

Important Links:

Go Back