Rethinking automatic chord recognition with convolutional neural networks

Eric J. Humphrey; Juan P. Bello

doi:10.1109/ICMLA.2012.220

Rethinking automatic chord recognition with convolutional neural networks

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

Abstract

Despite early success in automatic chord recognition, recent efforts are yielding diminishing returns while basically iterating over the same fundamental approach. Here, we abandon typical conventions and adopt a different perspective of the problem, where several seconds of pitch spectra are classified directly by a convolutional neural network. Using labeled data to train the system in a supervised manner, we achieve state of the art performance through this initial effort in an otherwise unexplored area. Subsequent error analysis provides insight into potential areas of improvement, and this approach to chord recognition shows promise for future harmonic analysis systems.

Original language	English (US)
Title of host publication	Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012
Pages	357-362
Number of pages	6
DOIs	https://doi.org/10.1109/ICMLA.2012.220
State	Published - 2012
Event	11th IEEE International Conference on Machine Learning and Applications, ICMLA 2012 - Boca Raton, FL, United States Duration: Dec 12 2012 → Dec 15 2012

Publication series

Name	Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012
Volume	2

Other

Other	11th IEEE International Conference on Machine Learning and Applications, ICMLA 2012
Country/Territory	United States
City	Boca Raton, FL
Period	12/12/12 → 12/15/12

Keywords

automatic music transcription
chord recognition
convolutional neural nets

ASJC Scopus subject areas

Human-Computer Interaction
Education

Access to Document

10.1109/ICMLA.2012.220

Cite this

Humphrey, E. J., & Bello, J. P. (2012). Rethinking automatic chord recognition with convolutional neural networks. In Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012 (pp. 357-362). Article 6406762 (Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012; Vol. 2). https://doi.org/10.1109/ICMLA.2012.220

Rethinking automatic chord recognition with convolutional neural networks. / Humphrey, Eric J.; Bello, Juan P.
Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012. 2012. p. 357-362 6406762 (Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012; Vol. 2).

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

Humphrey, EJ & Bello, JP 2012, Rethinking automatic chord recognition with convolutional neural networks. in Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012., 6406762, Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012, vol. 2, pp. 357-362, 11th IEEE International Conference on Machine Learning and Applications, ICMLA 2012, Boca Raton, FL, United States, 12/12/12. https://doi.org/10.1109/ICMLA.2012.220

@inproceedings{a55fd97165de4e56a65627637494b314,

title = "Rethinking automatic chord recognition with convolutional neural networks",

abstract = "Despite early success in automatic chord recognition, recent efforts are yielding diminishing returns while basically iterating over the same fundamental approach. Here, we abandon typical conventions and adopt a different perspective of the problem, where several seconds of pitch spectra are classified directly by a convolutional neural network. Using labeled data to train the system in a supervised manner, we achieve state of the art performance through this initial effort in an otherwise unexplored area. Subsequent error analysis provides insight into potential areas of improvement, and this approach to chord recognition shows promise for future harmonic analysis systems.",

keywords = "automatic music transcription, chord recognition, convolutional neural nets",

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

year = "2012",

doi = "10.1109/ICMLA.2012.220",

language = "English (US)",

isbn = "9780769549132",

series = "Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012",

pages = "357--362",

booktitle = "Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012",

note = "11th IEEE International Conference on Machine Learning and Applications, ICMLA 2012 ; Conference date: 12-12-2012 Through 15-12-2012",

}

TY - GEN

T1 - Rethinking automatic chord recognition with convolutional neural networks

AU - Humphrey, Eric J.

AU - Bello, Juan P.

PY - 2012

Y1 - 2012

N2 - Despite early success in automatic chord recognition, recent efforts are yielding diminishing returns while basically iterating over the same fundamental approach. Here, we abandon typical conventions and adopt a different perspective of the problem, where several seconds of pitch spectra are classified directly by a convolutional neural network. Using labeled data to train the system in a supervised manner, we achieve state of the art performance through this initial effort in an otherwise unexplored area. Subsequent error analysis provides insight into potential areas of improvement, and this approach to chord recognition shows promise for future harmonic analysis systems.

AB - Despite early success in automatic chord recognition, recent efforts are yielding diminishing returns while basically iterating over the same fundamental approach. Here, we abandon typical conventions and adopt a different perspective of the problem, where several seconds of pitch spectra are classified directly by a convolutional neural network. Using labeled data to train the system in a supervised manner, we achieve state of the art performance through this initial effort in an otherwise unexplored area. Subsequent error analysis provides insight into potential areas of improvement, and this approach to chord recognition shows promise for future harmonic analysis systems.

KW - automatic music transcription

KW - chord recognition

KW - convolutional neural nets

UR - http://www.scopus.com/inward/record.url?scp=84873577775&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84873577775&partnerID=8YFLogxK

U2 - 10.1109/ICMLA.2012.220

DO - 10.1109/ICMLA.2012.220

M3 - Conference contribution

AN - SCOPUS:84873577775

SN - 9780769549132

T3 - Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012

SP - 357

EP - 362

BT - Proceedings - 2012 11th International Conference on Machine Learning and Applications, ICMLA 2012

T2 - 11th IEEE International Conference on Machine Learning and Applications, ICMLA 2012

Y2 - 12 December 2012 through 15 December 2012

ER -

Rethinking automatic chord recognition with convolutional neural networks

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this