Key-independent classification of harmonic change in musical audio

Ernest Li, Juan Pablo Bello

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We introduce a novel method for describing the harmonic development of a musical signal by using only low-level audio features. Our approach uses Euclidean and phase distances in a Tonal Centroid space. Both measurements are taken between successive chroma partitions of a harmonically segmented signal, for each of three harmonic circles representing fifths, major thirds, and minor thirds. The resulting feature vector can be used to quantify a string of successive chord changes according to changes in chord quality and movement of the chordal root. We demonstrate that our feature set can provide both unique classification and accurate identification of harmonic changes, while resisting variations in orchestration and key.

Original languageEnglish (US)
Title of host publicationAudio Engineering Society - 123rd Audio Engineering Society Convention 2007
Pages1061-1068
Number of pages8
Volume2
StatePublished - 2007
Event123rd Audio Engineering Society Convention 2007 - New York, NY, United States
Duration: Oct 5 2007Oct 8 2007

Other

Other123rd Audio Engineering Society Convention 2007
CountryUnited States
CityNew York, NY
Period10/5/0710/8/07

Fingerprint

Harmonic
Chord or secant line
harmonics
Orchestration
Centroid
Feature Vector
centroids
Minor
partitions
Euclidean
Circle
Quantify
strings
Strings
Partition
Roots
Demonstrate
Movement

ASJC Scopus subject areas

  • Modeling and Simulation
  • Acoustics and Ultrasonics

Cite this

Li, E., & Bello, J. P. (2007). Key-independent classification of harmonic change in musical audio. In Audio Engineering Society - 123rd Audio Engineering Society Convention 2007 (Vol. 2, pp. 1061-1068)

Key-independent classification of harmonic change in musical audio. / Li, Ernest; Bello, Juan Pablo.

Audio Engineering Society - 123rd Audio Engineering Society Convention 2007. Vol. 2 2007. p. 1061-1068.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Li, E & Bello, JP 2007, Key-independent classification of harmonic change in musical audio. in Audio Engineering Society - 123rd Audio Engineering Society Convention 2007. vol. 2, pp. 1061-1068, 123rd Audio Engineering Society Convention 2007, New York, NY, United States, 10/5/07.
Li E, Bello JP. Key-independent classification of harmonic change in musical audio. In Audio Engineering Society - 123rd Audio Engineering Society Convention 2007. Vol. 2. 2007. p. 1061-1068
Li, Ernest ; Bello, Juan Pablo. / Key-independent classification of harmonic change in musical audio. Audio Engineering Society - 123rd Audio Engineering Society Convention 2007. Vol. 2 2007. pp. 1061-1068
@inproceedings{4b87378da48d40c5ba7d4bdd08b15f1d,
title = "Key-independent classification of harmonic change in musical audio",
abstract = "We introduce a novel method for describing the harmonic development of a musical signal by using only low-level audio features. Our approach uses Euclidean and phase distances in a Tonal Centroid space. Both measurements are taken between successive chroma partitions of a harmonically segmented signal, for each of three harmonic circles representing fifths, major thirds, and minor thirds. The resulting feature vector can be used to quantify a string of successive chord changes according to changes in chord quality and movement of the chordal root. We demonstrate that our feature set can provide both unique classification and accurate identification of harmonic changes, while resisting variations in orchestration and key.",
author = "Ernest Li and Bello, {Juan Pablo}",
year = "2007",
language = "English (US)",
isbn = "9781604239027",
volume = "2",
pages = "1061--1068",
booktitle = "Audio Engineering Society - 123rd Audio Engineering Society Convention 2007",

}

TY - GEN

T1 - Key-independent classification of harmonic change in musical audio

AU - Li, Ernest

AU - Bello, Juan Pablo

PY - 2007

Y1 - 2007

N2 - We introduce a novel method for describing the harmonic development of a musical signal by using only low-level audio features. Our approach uses Euclidean and phase distances in a Tonal Centroid space. Both measurements are taken between successive chroma partitions of a harmonically segmented signal, for each of three harmonic circles representing fifths, major thirds, and minor thirds. The resulting feature vector can be used to quantify a string of successive chord changes according to changes in chord quality and movement of the chordal root. We demonstrate that our feature set can provide both unique classification and accurate identification of harmonic changes, while resisting variations in orchestration and key.

AB - We introduce a novel method for describing the harmonic development of a musical signal by using only low-level audio features. Our approach uses Euclidean and phase distances in a Tonal Centroid space. Both measurements are taken between successive chroma partitions of a harmonically segmented signal, for each of three harmonic circles representing fifths, major thirds, and minor thirds. The resulting feature vector can be used to quantify a string of successive chord changes according to changes in chord quality and movement of the chordal root. We demonstrate that our feature set can provide both unique classification and accurate identification of harmonic changes, while resisting variations in orchestration and key.

UR - http://www.scopus.com/inward/record.url?scp=84866479265&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866479265&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781604239027

VL - 2

SP - 1061

EP - 1068

BT - Audio Engineering Society - 123rd Audio Engineering Society Convention 2007

ER -