Data-driven exploration of musical chord sequences

ABSTRACT

We present data-driven methods for supporting musical creativity by capturing the statistics of a musical database. Specifically, we introduce a system that supports users in exploring the high-dimensional space of musical chord sequences by parameterizing the variation among chord sequences in popular music. We provide a novel user interface that exposes these learned parameters as control axes, and we propose two automatic approaches for defining these axes. One approach is based on a novel clustering procedure, the other on principal components analysis. A user study compares our approaches for defining control axes both to each other and to an approach based on manually-assigned genre labels. Results show that our automatic methods for defining control axes provide a subjectively better user experience than axes based on manual genre labeling.

REFERENCES

• 1: Allan, M., Williams, C.K.I. Harmonising Chorales by Probabilistic Inference. Proc NIPS 2005.
• 2: Chuan, C.-H., Chew, E. A Hybrid System for Automatic Generation of Style--Specific Accompaniment. 4th Intl Joint Workshop on Computational Creativity, June 2007.
• 3: David Cope, Computers and musical style, A-R Editions, Inc., Madison, WI, 1991
• 4: Charles Dodge , Thomas A. Jerse, Computer Music: Synthesis, Composition and Performance, Macmillan Library Reference, 1997
• 5: A. K. Jain , M. N. Murty , P. J. Flynn, Data clustering: a review, ACM Computing Surveys (CSUR), v.31 n.3, p.264-323, Sept. 1999 [doi>10.1145/331499.331504]
• 6: Roberto Legaspi , Yuya Hashimoto , Koichi Moriyama , Satoshi Kurihara , Masayuki Numao, Music compositional intelligence with an affective flavor, Proceedings of the 12th international conference on Intelligent user interfaces, January 28-31, 2007, Honolulu, Hawaii, USA [doi>10.1145/1216295.1216335]
• 7: C. Karen Liu , Aaron Hertzmann , Zoran Popović, Learning physics-based motion style with nonlinear inverse optimization, ACM SIGGRAPH 2005 Papers, July 31-August 04, 2005, Los Angeles, California
• 8: Morris, D., Simon, I., and Basu, S. Exposing Parameters of a Trained Dynamic Model for Interactive Music Creation. Proc AAAI 2008.
• 9: PG Music Inc: Band-in-a-Box. http://pgmusic.com
• 10: Rohrmeier, M., and Cross, I.. Statistical Properties of Tonal Harmony in Bach's Chorales. Proc 10th Intl Conf on Music Perception and Cognition (ICMPC10). Sapporo, Japan, 2008.
• 11: Ian Simon , Dan Morris , Sumit Basu, MySong: automatic accompaniment generation for vocal melodies, Proceeding of the twenty-sixth annual SIGCHI conference on Human factors in computing systems, April 05-10, 2008, Florence, Italy [doi>10.1145/1357054.1357169]
• 12: Sleator, D. and Temperley, D. 2001. The Melisma Music Analyzer. http://www.link.cs.cmu.edu/melisma/
• 13: Wallis, I., Ingalls, T., and E. Campana. "Computer Generating Emotional Music: The Design of an Affective Music Algorithm." Proc Digital Audio Effects (DAFx-08). Espoo, Finland. 2008.

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.5 INFORMATION INTERFACES AND PRESENTATION (I.7)
      H.5.5 Sound and Music Computing
          Subjects: Modeling

Additional Classification:
  H. Information Systems
  H.5 INFORMATION INTERFACES AND PRESENTATION (I.7)
      H.5.5 Sound and Music Computing
          Subjects: Systems

General Terms:
Algorithms, Design, Human Factors

Keywords:
chords, clustering, creativity, genre, hmms, music, pca, transition matrix