Advertisement

The CEG Algorithm (Part II): Validation

  • Elaine ChewEmail author
Chapter
Part of the International Series in Operations Research & Management Science book series (ISOR, volume 204)

Abstract

In this chapter, I compare the performance of the Center of Effect Generator (CEG) key-finding method, based on the Spiral Array model, to that of two classic key-finding algorithms, namely, Krumhansl and Schmuckler’s Probe Tone Profile Method (PTPM), and Longuet-Higgins and Steedman’s Shape Matching Algorithm (SMA). The three algorithms are tested using the fugue subjects of Bach’s Well-Tempered Clavier; the evaluation criterion being the number of pitch events needed to correctly determine the key of the fugue. The CEG required on average 3.75 pitches, the PTPM 5.25, and the SMA 8.71. If one considered only the fugue subjects in which the tonic-dominant rule was not applied by any of the three algorithms, the CEG required 3.57 pitches, the PTPM 4.79, and the SMA 8.21. The MATLAB code and details of each test run are given in Appendix B. Here, I present an analysis of the results of the comparison, and discuss the issue of model validation. A MATLAB version of the CEG code appears in Appendix B. The CEG method is applied to polyphonic music in MIDI (Musical Instrument Digital Interface) format in Chap.  9, and adapted to music audio in Chap.  11.

Keywords

Minor Mode Scale Degree Pitch Class Pitch Collection Duration Information 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. 1.
    Bach, J.S.: 371 Harmonized Chorales and 69 Chorale Melodies with Figured Bass. (Revised, corrected, edited, and annotated by Riemenschneider, A.) G. Schirmer Inc., New York (1941)Google Scholar
  2. 2.
    Bach, J.S.: The Well-Tempered Clavier—Part I, BWV 846–869. Henle Urtext Edition, Munich (1997)Google Scholar
  3. 3.
    Bach, J.S.: The Well-Tempered Clavier—Part II, BWV 870–893. Henle Urtext Edition, Munich (2007)Google Scholar
  4. 4.
    Chuan, C.-H., Chew, E.: Creating ground truth for audio key-finding: when the title key may not be the key. In: Proceedings of the International Conference on Music Information Retrieval (2012)Google Scholar
  5. 5.
    Jonas, O.: Introduction to the Theory of Heinrich Shenker: the Nature of the Musical Work of Art. Longman, New York (1982)Google Scholar
  6. 6.
    Krumhansl, C.L., Kessler, E.J.: Tracing the dynamic changes in perceived tonal organisation in a spatial representation of musical keys. Psychol. Rev. 89(4), 334–368 (1982)Google Scholar
  7. 7.
    Krumhansl, C.L.: Cognitive Foundations of Musical Pitch. Oxford University Press, New York (1990)Google Scholar
  8. 8.
    Lerdahl, F., Jackendoff, R.: A Generative Theory of Tonal Music. MIT Press, Cambridge (1983)Google Scholar
  9. 9.
    Lewin, D.: Generalized Musical Intervals and Transformations. Yale University Press, New York (1987)Google Scholar
  10. 10.
    Longuet-Higgins, H.C., Steedman, M.J.: On interpreting Bach. In: Meltzer, B., Michie, D. (eds.) Machine Intelligence 6, pp. 221–241. Edinburgh University Press, Edinburgh (1971)Google Scholar
  11. 11.
    Temperley, D.: The Cognition of Basic Musical Structures. MIT Press, Cambridge (2001)Google Scholar

Copyright information

© Springer Science+Business Media New York 2014

Authors and Affiliations

  1. 1.Centre for Digital MusicQueen Mary University of LondonLondonUK

Personalised recommendations