BREVE: An HMPerceptron-Based Chord Recognition System

  • Daniele P. Radicioni
  • Roberto Esposito
Part of the Studies in Computational Intelligence book series (SCI, volume 274)


Tonal harmony analysis is a sophisticated task. It combines general knowledge with contextual cues, and it is concerned with faceted and evolving objects such as musical language, execution style and taste. We present Breve, a system for performing a particular kind of harmony analysis, chord recognition: music is encoded as a sequence of sounding events and the system should assing the appropriate chord label to each event. The solution proposed to the problem relies on a conditional model, where domain knowledge is encoded in the form of Boolean features. Breve exploits the recently proposed algorithm CarpeDiem to obtain significant computational gains in solving the optimization problem underlying the classification process. The implemented system has been validated on a corpus of chorales from J.S. Bach: we report and discuss the learnt weights, point out the committed errors, and elaborate on the correlation between errors and growth in the classification times in places where the music is less clearly asserted.


Chord Recognition Machine Learning Music Analysis 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Barthelemy, J., Bonardi, A.: Figured bass and tonality recognition. In: Procs. of the 2nd Annual International Symposium on Music Information Retrieval 2001 (2001)Google Scholar
  2. 2.
    Bent, I. (ed.): Music Analysis in the Nineteenth Century. Cambridge University Press, Cambridge (1994)Google Scholar
  3. 3.
    Bigand, E., Pineau, M.: Global context effects on musical expectancy. Perception and psychophysics 59(7), 1098–1107 (1997)Google Scholar
  4. 4.
    Collins, M.: Discriminative training methods for hidden markov models: Theory and experiments with perceptron algorithms. In: Proceedings of the Conference on Empirical Methods in Natural Language Processing (2002),
  5. 5.
    Cope, D.: A Musical Learning Algorithm. Comput. Music J. 28(3), 12–27 (2004)CrossRefGoogle Scholar
  6. 6.
    Demany, L., Semal, C.: Harmonic and melodic octave templates. The Journal of the Acoustical Society of America 88(5), 2126–2135 (1990)CrossRefGoogle Scholar
  7. 7.
    Dietterich, T.: Machine Learning for Sequential Data: A Review. In: Caelli, T.M., Amin, A., Duin, R.P.W., Kamel, M.S., de Ridder, D. (eds.) SPR 2002 and SSPR 2002. LNCS, vol. 2396, pp. 15–30. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  8. 8.
    Dietterich, T.G., Ashenfelter, A., Bulatov, Y.: Training conditional random fields via gradient tree boosting. In: ICML 2004: Twenty-first international conference on Machine learning. ACM Press, New York (2004), Google Scholar
  9. 9.
    Esposito, R., Radicioni, D.P.: CarpeDiem: an Algorithm for the Fast Evaluation of SSL Classifiers. In: Proceedings of the 24th Annual International Conference on Machine Learning, ICML 2007 (2007)Google Scholar
  10. 10.
    Esposito, R., Radicioni, D.P.: Trip Around the HMPerceptron Algorithm: Empirical Findings and Theoretical Tenets. In: Basili, R., Pazienza, M.T. (eds.) AI*IA 2007. LNCS (LNAI), vol. 4733, pp. 242–253. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  11. 11.
    Freeman, J.: Fast generation of audio signatures to describe iTunes libraries. Journal of New Music Research 35(1), 51–61 (2006)CrossRefGoogle Scholar
  12. 12.
    Grachten, M., Arcos, J.L., López de Mantaras, R.: A Case Based Approach to Expressivity-Aware Tempo Transformation. Mach. Learn. 65(2-3), 411–437 (2006)CrossRefGoogle Scholar
  13. 13.
    Harte, C., Sandler, M., Abdallah, S., Gomez, E.: Symbolic Representation of Musical Chords: A Proposed Syntax for Text Annotation. In: International Conference on Music Information Retrieval, pp. 66–71. Queen Mary, University of London (2005)Google Scholar
  14. 14.
    Kostka, S., Payne, D.: Tonal Harmony. McGraw-Hill, New York (1984)Google Scholar
  15. 15.
    Lafferty, J., Pereira, F.: Conditional random fields: Probabilistic models for segmenting and labeling sequence data. In: Procs. of the 18th International Conference on Machine Learning, pp. 282–289. Morgan Kaufmann, San Francisco (2001)Google Scholar
  16. 16.
    Lee, K., Slaney, M.: Acoustic Chord Transcription and Key Extraction from Audio Using Key-Dependent HMMs Trained on Synthesized Audio. IEEE Transactions on Audio, Speech and Language Processing 16, 291–301 (2008)CrossRefGoogle Scholar
  17. 17.
    Lerdahl, F., Jackendoff, R.: A Generative Theory of Tonal Music. MIT Press, Cambridge (1983)Google Scholar
  18. 18.
    López de Mantaras, R., Arcos, J.: AI and Music: From Composition to Expressive Performance. AI Magazine 23, 43–57 (2002)Google Scholar
  19. 19.
    McCallum, A., Freitag, D., Pereira, F.: Maximum entropy Markov models for information extraction and segmentation. In: Proc. 17th International Conf. on Machine Learning, pp. 591–598. Morgan Kaufmann, San Francisco (2000), Google Scholar
  20. 20.
    Pardo, B., Birmingham, W.P.: Automated partitioning of tonal music. In: Procs. of the Thirteenth International Florida Artificial Intelligence Research Society Conference, FLAIRS 2000 (2000)Google Scholar
  21. 21.
    Pardo, B., Birmingham, W.P.: Algorithms for chordal analysis. Comput. Music J. 26, 27–49 (2002)CrossRefGoogle Scholar
  22. 22.
    Rabiner, L.R.: A tutorial on Hidden Markov Models and Selected Applications in Speech Recognition. Proceedings of the IEEE 77, 267–296 (1989)CrossRefGoogle Scholar
  23. 23.
    Raphael, C.: A Hybrid Graphical Model for Rhytmic Parsing. Artif. Intell. 137(1-2), 217–238 (2002)CrossRefzbMATHGoogle Scholar
  24. 24.
    Raphael, C., Stoddard, J.: Functional harmonic analysis using probabilistic models. Computer Music Journal 28(3), 45–52 (2004)CrossRefGoogle Scholar
  25. 25.
    Rosenblatt, F.: The perceptron: A probabilistic model for information storage and organization in the brain. Psychological Review (Reprinted in Neurocomputing (MIT Press, 1998)) 65, 386–408 (1958)MathSciNetGoogle Scholar
  26. 26.
    Schoenberg, A.: Structural Functions in Harmony. Norton, New York (1969)Google Scholar
  27. 27.
    Scholz, R., Ramalho, G.: Cochonut: Recognizing Complex Chords from MIDI Guitar Sequences. In: Procs. of the 2008 International Conference on Music Information Retrieval, pp. 27–32 (2008)Google Scholar
  28. 28.
    Temperley, D.: The Cognition of Basic Musical Structures. MIT Press, Cambridge (2001)Google Scholar
  29. 29.
    Viterbi, A.J.: Error Bounds for Convolutional Codes and an Asymptotically Optimum Decoding Algorithm. IEEE Transactions on Information Theory 13, 260–269 (1967)CrossRefzbMATHGoogle Scholar
  30. 30.
    Widmer, G.: Discovering Simple Rules in Complex Data: A Meta-learning Algorithm and Some Surprising Musical Discoveries. Artif. Intell. 146, 129–148 (2001)CrossRefMathSciNetGoogle Scholar
  31. 31.
    Widmer, G.: Guest editorial: Machine learning in and for music. Machine Learning 65, 343–346 (2006)CrossRefGoogle Scholar
  32. 32.
    Winograd, T.: Linguistics and computer analysis of tonal harmony. Journal of Music Theory 12, 2–49 (1968)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Daniele P. Radicioni
    • 1
  • Roberto Esposito
    • 1
  1. 1.Dipartimento di InformaticaUniversità di TorinoTorino

Personalised recommendations