Dynamic Change Points in Music Audio Capture Dynamic Markings in Score

Katerina Kosta*, Rebecca Killick, Oscar Bandtlow, Elaine Chew

*Corresponding author for this work

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

10 Downloads (Pure)

Abstract

We present an analysis of change-points in the loudness time series of audio recordings of Chopin’s Mazurkas. The change-point positions are derived using the PELT algo- rithm and a heuristic is introduced for the automatic ex- traction of change points of multiple recordings of the same piece. We map the change points to the correspond- ing score position for analysis. The results show that sig- nificant dynamic score markings do indeed correspond to change points; and, evidence suggests that change points in score positions without dynamic markings correspond to structurally salient events or events based on temporal changes.
Original languageEnglish
Title of host publicationInternational Society for Music Information Retrieval Conference
Number of pages2
Publication statusPublished - 2017
Event18th International Society for Music Information Retrieval Conference - National University of Singapore Research Institute (NUSRI), Suzhou, China
Duration: 23 Oct 201727 Oct 2017
Conference number: 18
https://ismir2017.ismir.net/

Conference

Conference18th International Society for Music Information Retrieval Conference
Abbreviated titleISMIR 2017
Country/TerritoryChina
CitySuzhou
Period23/10/201727/10/2017
Internet address

Keywords

  • changepoints
  • statistics
  • loudness
  • dynamics
  • music expression
  • expressivity
  • prosody

Fingerprint

Dive into the research topics of 'Dynamic Change Points in Music Audio Capture Dynamic Markings in Score'. Together they form a unique fingerprint.

Cite this