Learning a robust Tonnetz-space transform for automatic chord recognition

Eric J. Humphrey; Taemin Cho; Juan P. Bello

doi:10.1109/ICASSP.2012.6287914

Learning a robust Tonnetz-space transform for automatic chord recognition

Eric J. Humphrey, Taemin Cho, Juan P. Bello

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Temporal pitch class profiles - commonly referred to as a chromagrams - are the de facto standard signal representation for content-based methods of musical harmonic analysis, despite exhibiting a set of practical difficulties. Here, we present a novel, data-driven approach to learning a robust function that projects audio data into Tonnetz-space, a geometric representation of equal-tempered pitch intervals grounded in music theory. We apply this representation to automatic chord recognition and show that our approach out-performs the classification accuracy of previous chroma representations, while providing a mid-level feature space that circumvents challenges inherent to chroma.

Original language	English (US)
Title of host publication	2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings
Pages	453-456
Number of pages	4
DOIs	https://doi.org/10.1109/ICASSP.2012.6287914
State	Published - 2012
Event	2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Kyoto, Japan Duration: Mar 25 2012 → Mar 30 2012

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ISSN (Print)	1520-6149

Other

Other	2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012
Country/Territory	Japan
City	Kyoto
Period	3/25/12 → 3/30/12

Keywords

Chord Recognition
Convolutional Neural Networks
Deep Learning
Tonnetz

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/ICASSP.2012.6287914

Cite this

Humphrey, E. J., Cho, T., & Bello, J. P. (2012). Learning a robust Tonnetz-space transform for automatic chord recognition. In 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings (pp. 453-456). Article 6287914 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). https://doi.org/10.1109/ICASSP.2012.6287914

Learning a robust Tonnetz-space transform for automatic chord recognition. / Humphrey, Eric J.; Cho, Taemin; Bello, Juan P.
2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings. 2012. p. 453-456 6287914 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Humphrey, EJ, Cho, T & Bello, JP 2012, Learning a robust Tonnetz-space transform for automatic chord recognition. in 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings., 6287914, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, pp. 453-456, 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012, Kyoto, Japan, 3/25/12. https://doi.org/10.1109/ICASSP.2012.6287914

Humphrey EJ, Cho T, Bello JP. Learning a robust Tonnetz-space transform for automatic chord recognition. In 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings. 2012. p. 453-456. 6287914. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP.2012.6287914

@inproceedings{c4b1ac7f28e04d319996913bc4cfdc4d,

title = "Learning a robust Tonnetz-space transform for automatic chord recognition",

abstract = "Temporal pitch class profiles - commonly referred to as a chromagrams - are the de facto standard signal representation for content-based methods of musical harmonic analysis, despite exhibiting a set of practical difficulties. Here, we present a novel, data-driven approach to learning a robust function that projects audio data into Tonnetz-space, a geometric representation of equal-tempered pitch intervals grounded in music theory. We apply this representation to automatic chord recognition and show that our approach out-performs the classification accuracy of previous chroma representations, while providing a mid-level feature space that circumvents challenges inherent to chroma.",

keywords = "Chord Recognition, Convolutional Neural Networks, Deep Learning, Tonnetz",

author = "Humphrey, {Eric J.} and Taemin Cho and Bello, {Juan P.}",

year = "2012",

doi = "10.1109/ICASSP.2012.6287914",

language = "English (US)",

isbn = "9781467300469",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

pages = "453--456",

booktitle = "2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings",

note = "2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 ; Conference date: 25-03-2012 Through 30-03-2012",

}

TY - GEN

T1 - Learning a robust Tonnetz-space transform for automatic chord recognition

AU - Humphrey, Eric J.

AU - Cho, Taemin

AU - Bello, Juan P.

PY - 2012

Y1 - 2012

N2 - Temporal pitch class profiles - commonly referred to as a chromagrams - are the de facto standard signal representation for content-based methods of musical harmonic analysis, despite exhibiting a set of practical difficulties. Here, we present a novel, data-driven approach to learning a robust function that projects audio data into Tonnetz-space, a geometric representation of equal-tempered pitch intervals grounded in music theory. We apply this representation to automatic chord recognition and show that our approach out-performs the classification accuracy of previous chroma representations, while providing a mid-level feature space that circumvents challenges inherent to chroma.

AB - Temporal pitch class profiles - commonly referred to as a chromagrams - are the de facto standard signal representation for content-based methods of musical harmonic analysis, despite exhibiting a set of practical difficulties. Here, we present a novel, data-driven approach to learning a robust function that projects audio data into Tonnetz-space, a geometric representation of equal-tempered pitch intervals grounded in music theory. We apply this representation to automatic chord recognition and show that our approach out-performs the classification accuracy of previous chroma representations, while providing a mid-level feature space that circumvents challenges inherent to chroma.

KW - Chord Recognition

KW - Convolutional Neural Networks

KW - Deep Learning

KW - Tonnetz

UR - http://www.scopus.com/inward/record.url?scp=84867602860&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867602860&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2012.6287914

DO - 10.1109/ICASSP.2012.6287914

M3 - Conference contribution

AN - SCOPUS:84867602860

SN - 9781467300469

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 453

EP - 456

BT - 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings

T2 - 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012

Y2 - 25 March 2012 through 30 March 2012

ER -

Learning a robust Tonnetz-space transform for automatic chord recognition

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this