Audio properties of perceived boundaries in music

Jordan B.L. Smith, Ching-Hua Chuan, Elaine Chew

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Data mining tasks such as music indexing, information retrieval, and similarity search, require an understanding of how listeners process music internally. Many algorithms for automatically analyzing the structure of recorded music assume that a large change in one or another musical feature suggests a section boundary. However, this assumption has not been tested: while our understanding of how listeners segment melodies has advanced greatly in the past decades, little is known about how this process works with more complex, full-textured pieces of music, or how stable this process is across genres. Knowing how these factors affect how boundaries are perceived will help researchers to judge the viability of algorithmic approaches with different corpora of music. We present a statistical analysis of a large corpus of recordings whose formal structure was annotated by expert listeners. We find that the acoustic properties of boundaries in these recordings corroborate findings of previous perceptual experiments. Nearly all boundaries correspond to peaks in novelty functions, which measure the rate of change of a musical feature at a particular time scale. Moreover, most of these boundaries match peaks in novelty for several features at several time scales. We observe that the boundary-novelty relationship can vary with listener, time scale, genre, and musical feature. Finally, we show that a boundary profile derived from a collection of novelty functions correlates with the estimated salience of boundaries indicated by listeners.

Original languageEnglish (US)
Article number6856249
Pages (from-to)1219-1228
Number of pages10
JournalIEEE Transactions on Multimedia
Volume16
Issue number5
DOIs
StatePublished - Jan 1 2014
Externally publishedYes

Fingerprint

Acoustic properties
Information retrieval
Data mining
Statistical methods
Experiments

Keywords

  • Boundaries
  • corpus analysis
  • music analysis
  • music information retrieval

ASJC Scopus subject areas

  • Signal Processing
  • Media Technology
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Audio properties of perceived boundaries in music. / Smith, Jordan B.L.; Chuan, Ching-Hua; Chew, Elaine.

In: IEEE Transactions on Multimedia, Vol. 16, No. 5, 6856249, 01.01.2014, p. 1219-1228.

Research output: Contribution to journalArticle

Smith, Jordan B.L. ; Chuan, Ching-Hua ; Chew, Elaine. / Audio properties of perceived boundaries in music. In: IEEE Transactions on Multimedia. 2014 ; Vol. 16, No. 5. pp. 1219-1228.
@article{acb3773a9df1458aa052df8baebf4ea7,
title = "Audio properties of perceived boundaries in music",
abstract = "Data mining tasks such as music indexing, information retrieval, and similarity search, require an understanding of how listeners process music internally. Many algorithms for automatically analyzing the structure of recorded music assume that a large change in one or another musical feature suggests a section boundary. However, this assumption has not been tested: while our understanding of how listeners segment melodies has advanced greatly in the past decades, little is known about how this process works with more complex, full-textured pieces of music, or how stable this process is across genres. Knowing how these factors affect how boundaries are perceived will help researchers to judge the viability of algorithmic approaches with different corpora of music. We present a statistical analysis of a large corpus of recordings whose formal structure was annotated by expert listeners. We find that the acoustic properties of boundaries in these recordings corroborate findings of previous perceptual experiments. Nearly all boundaries correspond to peaks in novelty functions, which measure the rate of change of a musical feature at a particular time scale. Moreover, most of these boundaries match peaks in novelty for several features at several time scales. We observe that the boundary-novelty relationship can vary with listener, time scale, genre, and musical feature. Finally, we show that a boundary profile derived from a collection of novelty functions correlates with the estimated salience of boundaries indicated by listeners.",
keywords = "Boundaries, corpus analysis, music analysis, music information retrieval",
author = "Smith, {Jordan B.L.} and Ching-Hua Chuan and Elaine Chew",
year = "2014",
month = "1",
day = "1",
doi = "10.1109/TMM.2014.2310706",
language = "English (US)",
volume = "16",
pages = "1219--1228",
journal = "IEEE Transactions on Multimedia",
issn = "1520-9210",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

TY - JOUR

T1 - Audio properties of perceived boundaries in music

AU - Smith, Jordan B.L.

AU - Chuan, Ching-Hua

AU - Chew, Elaine

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Data mining tasks such as music indexing, information retrieval, and similarity search, require an understanding of how listeners process music internally. Many algorithms for automatically analyzing the structure of recorded music assume that a large change in one or another musical feature suggests a section boundary. However, this assumption has not been tested: while our understanding of how listeners segment melodies has advanced greatly in the past decades, little is known about how this process works with more complex, full-textured pieces of music, or how stable this process is across genres. Knowing how these factors affect how boundaries are perceived will help researchers to judge the viability of algorithmic approaches with different corpora of music. We present a statistical analysis of a large corpus of recordings whose formal structure was annotated by expert listeners. We find that the acoustic properties of boundaries in these recordings corroborate findings of previous perceptual experiments. Nearly all boundaries correspond to peaks in novelty functions, which measure the rate of change of a musical feature at a particular time scale. Moreover, most of these boundaries match peaks in novelty for several features at several time scales. We observe that the boundary-novelty relationship can vary with listener, time scale, genre, and musical feature. Finally, we show that a boundary profile derived from a collection of novelty functions correlates with the estimated salience of boundaries indicated by listeners.

AB - Data mining tasks such as music indexing, information retrieval, and similarity search, require an understanding of how listeners process music internally. Many algorithms for automatically analyzing the structure of recorded music assume that a large change in one or another musical feature suggests a section boundary. However, this assumption has not been tested: while our understanding of how listeners segment melodies has advanced greatly in the past decades, little is known about how this process works with more complex, full-textured pieces of music, or how stable this process is across genres. Knowing how these factors affect how boundaries are perceived will help researchers to judge the viability of algorithmic approaches with different corpora of music. We present a statistical analysis of a large corpus of recordings whose formal structure was annotated by expert listeners. We find that the acoustic properties of boundaries in these recordings corroborate findings of previous perceptual experiments. Nearly all boundaries correspond to peaks in novelty functions, which measure the rate of change of a musical feature at a particular time scale. Moreover, most of these boundaries match peaks in novelty for several features at several time scales. We observe that the boundary-novelty relationship can vary with listener, time scale, genre, and musical feature. Finally, we show that a boundary profile derived from a collection of novelty functions correlates with the estimated salience of boundaries indicated by listeners.

KW - Boundaries

KW - corpus analysis

KW - music analysis

KW - music information retrieval

UR - http://www.scopus.com/inward/record.url?scp=84904758556&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84904758556&partnerID=8YFLogxK

U2 - 10.1109/TMM.2014.2310706

DO - 10.1109/TMM.2014.2310706

M3 - Article

AN - SCOPUS:84904758556

VL - 16

SP - 1219

EP - 1228

JO - IEEE Transactions on Multimedia

JF - IEEE Transactions on Multimedia

SN - 1520-9210

IS - 5

M1 - 6856249

ER -