Music pattern discovery with variable Markov oracle: A unified approach to symbolic and audio representations

Cheng i. Wang, Jennifer Hsu, Shlomo Dubnov

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

12 Scopus citations

Abstract

This paper presents a framework for automatically discovering patterns in a polyphonic music piece. The proposed framework is capable of handling both symbolic and audio representations. Chroma features are post-processed with heuristics stemming from musical knowledge and fed into the pattern discovery framework. The pattern-finding algorithm is based on Variable Markov Oracle. The Variable Markov Oracle data structure is capable of locating repeated suffixes within a time series, thus making it an appropriate tool for the pattern discovery task. Evaluation of the proposed framework is performed on the JKU Patterns Development Dataset with state of the art performance.

Original languageEnglish
Title of host publicationProceedings of the 16th International Society for Music Information Retrieval Conference, ISMIR 2015
EditorsMeinard Muller, Frans Wiering
PublisherInternational Society for Music Information Retrieval
Pages176-182
Number of pages7
ISBN (Electronic)9788460688532
StatePublished - 1 Jan 2015
Externally publishedYes
Event16th International Society for Music Information Retrieval Conference, ISMIR 2015 - Malaga, Spain
Duration: 26 Oct 201530 Oct 2015

Publication series

NameProceedings of the 16th International Society for Music Information Retrieval Conference, ISMIR 2015

Conference

Conference16th International Society for Music Information Retrieval Conference, ISMIR 2015
Country/TerritorySpain
CityMalaga
Period26/10/1530/10/15

ASJC Scopus subject areas

  • Music
  • Information Systems

Fingerprint

Dive into the research topics of 'Music pattern discovery with variable Markov oracle: A unified approach to symbolic and audio representations'. Together they form a unique fingerprint.

Cite this