Advanced search

Using online chord databases to ehance chord recognition

(2011) JOURNAL OF NEW MUSIC RESEARCH. 40(2). p.139-152
Author
Organization
Abstract
Advances in chord recognition research using machine learning are hampered by two factors: the scarcity of annotated training data, and the limited complexity of the features and models used. Both problems are intertwined, as with few training examples, increasing the complexity of the model would inevitably lead to overfitting. In this paper we develop a way to address the first problem by exploiting chord annotations from online chord databases. We show how such chord annotations, despite being noisy and lacking exact chord onset times, can be put to use both during the recognition and training stage. We note that the ability to exploit this large untapped resource may enable researchers to also address the second problem: with more training data, one may be able to use more complex models without running the same high risk of overfitting.
Keywords
AUDIO

Citation

Please use this url to cite or link to this publication:

Chicago
McVicar, Matt, Yizhao Ni, Raul Santos-Rodriguez, and Tijl De Bie. 2011. “Using Online Chord Databases to Ehance Chord Recognition.” Journal of New Music Research 40 (2): 139–152.
APA
McVicar, M., Ni, Y., Santos-Rodriguez, R., & De Bie, T. (2011). Using online chord databases to ehance chord recognition. JOURNAL OF NEW MUSIC RESEARCH, 40(2), 139–152.
Vancouver
1.
McVicar M, Ni Y, Santos-Rodriguez R, De Bie T. Using online chord databases to ehance chord recognition. JOURNAL OF NEW MUSIC RESEARCH. 2011;40(2):139–52.
MLA
McVicar, Matt et al. “Using Online Chord Databases to Ehance Chord Recognition.” JOURNAL OF NEW MUSIC RESEARCH 40.2 (2011): 139–152. Print.
@article{6936413,
  abstract     = {Advances in chord recognition research using machine learning are hampered by two factors: the scarcity of annotated training data, and the limited complexity of the features and models used. Both problems are intertwined, as with few training examples, increasing the complexity of the model would inevitably lead to overfitting. In this paper we develop a way to address the first problem by exploiting chord annotations from online chord databases. We show how such chord annotations, despite being noisy and lacking exact chord onset times, can be put to use both during the recognition and training stage. We note that the ability to exploit this large untapped resource may enable researchers to also address the second problem: with more training data, one may be able to use more complex models without running the same high risk of overfitting.},
  author       = {McVicar, Matt and Ni, Yizhao and Santos-Rodriguez, Raul and De Bie, Tijl},
  issn         = {0929-8215},
  journal      = {JOURNAL OF NEW MUSIC RESEARCH},
  keywords     = {AUDIO},
  language     = {eng},
  number       = {2},
  pages        = {139--152},
  title        = {Using online chord databases to ehance chord recognition},
  url          = {http://dx.doi.org/10.1080/09298215.2011.573564},
  volume       = {40},
  year         = {2011},
}

Altmetric
View in Altmetric
Web of Science
Times cited: