Automatic Chord estimation on seventhsbass Chord vocabulary using deep neural network

Junqi Deng; Yu Kwong Kwok

doi:10.1109/ICASSP.2016.7471677

Automatic Chord estimation on seventhsbass Chord vocabulary using deep neural network

Junqi Deng
, Yu Kwong Kwok

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

8 Citations (Scopus)

Abstract

This paper proposes an automatic chord estimation (ACE) system with a two-layer architecture. The first layer performs chord smoothing with «GMM + HMM» approach. Then given the results of the first layer, the second layer performs chord estimation using a deep neural network, which is trained on a well chord-type balanced dataset. The system accepts exactly the «SeventhsBass» vocabulary. Three approaches with different configurations of the system are compared with Chordino, which is probably the only both MIREX evaluated and «SeventhsBass» acceptable ACE system. Evaluation results on «The Beatles» dataset show that the best approach outperforms Chordino in the most difficult «SeventhsBass» metric in a significant way.

Original language	English
Title of host publication	2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings
Pages	261-265
Number of pages	5
ISBN (Electronic)	9781479999880
DOIs	https://doi.org/10.1109/ICASSP.2016.7471677
Publication status	Published - 18 May 2016
Externally published	Yes
Event	41st IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Shanghai, China Duration: 20 Mar 2016 → 25 Mar 2016

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	2016-May
ISSN (Print)	1520-6149

Conference

Conference	41st IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016
Country/Territory	China
City	Shanghai
Period	20/03/16 → 25/03/16

Keywords

Automatic Chord Estimation
Deep Belief Network
Deep Neural Network

Access to Document

10.1109/ICASSP.2016.7471677

Cite this

Deng, J., & Kwok, Y. K. (2016). Automatic Chord estimation on seventhsbass Chord vocabulary using deep neural network. In 2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings (pp. 261-265). Article 7471677 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2016-May). https://doi.org/10.1109/ICASSP.2016.7471677

@inproceedings{4ee80e67ce4648a7b3610e30ca8babb3,

title = "Automatic Chord estimation on seventhsbass Chord vocabulary using deep neural network",

abstract = "This paper proposes an automatic chord estimation (ACE) system with a two-layer architecture. The first layer performs chord smoothing with «GMM + HMM» approach. Then given the results of the first layer, the second layer performs chord estimation using a deep neural network, which is trained on a well chord-type balanced dataset. The system accepts exactly the «SeventhsBass» vocabulary. Three approaches with different configurations of the system are compared with Chordino, which is probably the only both MIREX evaluated and «SeventhsBass» acceptable ACE system. Evaluation results on «The Beatles» dataset show that the best approach outperforms Chordino in the most difficult «SeventhsBass» metric in a significant way.",

keywords = "Automatic Chord Estimation, Deep Belief Network, Deep Neural Network",

author = "Junqi Deng and Kwok, \{Yu Kwong\}",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 41st IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 ; Conference date: 20-03-2016 Through 25-03-2016",

year = "2016",

month = may,

day = "18",

doi = "10.1109/ICASSP.2016.7471677",

language = "English",

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

pages = "261--265",

booktitle = "2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings",

}

Deng, J & Kwok, YK 2016, Automatic Chord estimation on seventhsbass Chord vocabulary using deep neural network. in 2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings., 7471677, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 2016-May, pp. 261-265, 41st IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016, Shanghai, China, 20/03/16. https://doi.org/10.1109/ICASSP.2016.7471677

Automatic Chord estimation on seventhsbass Chord vocabulary using deep neural network. / Deng, Junqi; Kwok, Yu Kwong.
2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings. 2016. p. 261-265 7471677 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2016-May).

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

TY - GEN

T1 - Automatic Chord estimation on seventhsbass Chord vocabulary using deep neural network

AU - Deng, Junqi

AU - Kwok, Yu Kwong

PY - 2016/5/18

Y1 - 2016/5/18

N2 - This paper proposes an automatic chord estimation (ACE) system with a two-layer architecture. The first layer performs chord smoothing with «GMM + HMM» approach. Then given the results of the first layer, the second layer performs chord estimation using a deep neural network, which is trained on a well chord-type balanced dataset. The system accepts exactly the «SeventhsBass» vocabulary. Three approaches with different configurations of the system are compared with Chordino, which is probably the only both MIREX evaluated and «SeventhsBass» acceptable ACE system. Evaluation results on «The Beatles» dataset show that the best approach outperforms Chordino in the most difficult «SeventhsBass» metric in a significant way.

AB - This paper proposes an automatic chord estimation (ACE) system with a two-layer architecture. The first layer performs chord smoothing with «GMM + HMM» approach. Then given the results of the first layer, the second layer performs chord estimation using a deep neural network, which is trained on a well chord-type balanced dataset. The system accepts exactly the «SeventhsBass» vocabulary. Three approaches with different configurations of the system are compared with Chordino, which is probably the only both MIREX evaluated and «SeventhsBass» acceptable ACE system. Evaluation results on «The Beatles» dataset show that the best approach outperforms Chordino in the most difficult «SeventhsBass» metric in a significant way.

KW - Automatic Chord Estimation

KW - Deep Belief Network

KW - Deep Neural Network

UR - https://www.scopus.com/pages/publications/84973299046

U2 - 10.1109/ICASSP.2016.7471677

DO - 10.1109/ICASSP.2016.7471677

M3 - Conference contribution

AN - SCOPUS:84973299046

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

SP - 261

EP - 265

BT - 2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings

T2 - 41st IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016

Y2 - 20 March 2016 through 25 March 2016

ER -

Deng J, Kwok YK. Automatic Chord estimation on seventhsbass Chord vocabulary using deep neural network. In 2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings. 2016. p. 261-265. 7471677. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP.2016.7471677

Automatic Chord estimation on seventhsbass Chord vocabulary using deep neural network

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this