Abstract
In this chapter, we discuss how to render expressive polyphonic piano music through a statistical approach. Generating polyphonic expression is an important element in achieving automatic expressive piano performance since the piano is a polyphonic instrument. We will start by discussing the features of polyphonic piano expression and present a method for modeling it based on an approximation involving melodies and harmonies. An experimental evaluation indicates that performances generated with the proposed method achieved polyphonic expression and created an impression of expressiveness. In addition, performances generated with models trained on different performances were perceptually distinguishable by human listeners. Finally, we introduce an automatic expressive piano system called Polyhymnia that won the first place in the autonomous section of Performance Rendering Contest for Computer Systems (RenCon) 2010.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
Six nonmusicians, 17 hobbyist musicians, and 2 professional musicians participated in the experiment.
- 2.
Differences in the average scores were tested using analysis of variance and its post-hoc test using TukeyHSD algorithm implemented in GNU R.
- 3.
Polyhymnia is one of the nine muses in Greek mythology, and it also means choral poetry or multiple hymns.
- 4.
Two professional musicians, 13 hobbyist musicians, and 2 nonmusicians participated in the listening experiments.
- 5.
F. Chopin, Prelude No. ~ 1, 4, 7, 15, 20, Etude Op. ~ 10-3, 10–4, 25–11, Waltz Op. ~ 18, 34–2, 64–2, 69–1, 69–2, Nocturne No. ~ 2 and 10.
- 6.
W. A. Mozart, Piano Sonata KV279-1, 279–2, 279–3, 331–1, 545–1, 545–2 and 545-3.
- 7.
W. A. Mozart, Klavierstück, KV. 33, W. A. Mozart, Sonatina No. 41, KV. 439, F. Chopin, Mazurka No. 19, Op. 30.
References
Berger AL (1997) The improved iterative scaling algorithm: a gentle introduction, Carnegie Mellon University
Bottou L (1991) Stochastic gradient learning in neural networks. In Proceedings of Neuro-Nimes 91
Darroch J, Ratcliff D (1972) Generalized iterative scaling for log-linear models. Ann Math Stat 43:1470–1480
Flossmann S, Grachten M, Widmer G (2009) Expressive performance rendering: introducing performance context. In Proceedings of the SMC, pp 155–160
Friberg A, Bresin R, Sundberg J (2006) Overview of the KTH rule system for musical performance. Adv Cogn Psychol 2(2):145–161
Gabrielsson A (1985) Interplay between analysis and synthesis in studies of music performance and music experience. Music Percept 3:59–86
Gieseking A, Leimer K (1972) Piano technique. Dover, New York
Grindlay G, Helmhold D (2006) Modeling, analyzing, and synthesizing expressive piano performance with graphical models. Mach Learn 65:361–387
Hashida M, Matsui T, Katayose H (2008) A new database describing deviation information of performance expressions. In Proceedings of the ISMIR, pp 489–494
Huron D, Fantini D (1989) The avoidance of inner-voice entries: perceptual evidence and musical practice. Music Percept 7(1):43–48
Lafferty J, McCallum A, Pereira F (2011) Conditional random fields: probabilistic models for segmenting and labeling sequence data. In Proceedings of the ML, pp 282–289
Lehvinne J (1972) Basic principle in pianoforte playing. Dover, New York
Lerdahl F, Jackendoff R (1983) A generative theory of tonal music. MIT Press, Cambridge, MA
Mazzola G (2002) The topos of music – geometric logic of concept, theory, and performance. Birkenhäuser, Basel/Boston
MusicXML, Recordare llc., http://www.recordare.com/musicxml
Palmer C (1997) Music performance. Ann Rev Psychol 48:115–138
Pancutt R (2003) Accents and expression in piano performance. In: Niemöller KW (ed) Perspektiven und Methoden einer Systemischen Musikwissenschaft, Systemische Musikwissenschaft. Peter Lang, Frankfurt am Main, pp 163–185
Pietra SD, Pietra VD, Lafferty J (1995) Inducing features of random fields. Technical Report, CMU-CS-95–144, Carnegie Mellon University
Repp BH (1992) Diversity and commonality in music performance: an analysis of timing microstructure in Schumann’s “Trämerei”. J Acoust Soc Am 92(5):2546–2568
Sha F, Pereira F (2003) Shallow parsing with conditional random fields. In Proceedings of Human Language Technology, NAACL
Sundberg J, Askenfelt A, Frydèn L (1983) Musical performance: a synthesis-by-rule approach. Comp Music J 7(1):37–43
Suzuki T, Tokunaga T, Tanaka H (1999) Case-based approach to the generation of musical expression. In Proceedings of the IJCAI, pp 642–648
Teramura K, Okuma H (2008) Gaussian process regression for rendering music performance. In Proceedings of the ICMPC
Teramura K, Maeda S (2010) Statistical learning of tempo variation for imitating piano performance (in Japanese). IPSJ Tech Rep., vol. 85, no. 12
Wallach HM (2002) Efficient training of conditional random fields. Master’s thesis, University of Edinburgh
Widmer G, Dixon S, Goebl W, Pampalk E, Tobudic A (2003) In search of the Horowitz factor. AI Mag 24(3):111–130
Acknowledgments
This work was partially funded by CrestMuse Project of Japan Science and Technology Agency and supported by Samsung Scholarship Foundation.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Questions
Questions
-
1.
What is one of the key factors in the expressive ability of the piano?
-
2.
Name three benefits of the statistical modeling approach for modeling expressive performance.
-
3.
Highlight four features of musical expression in polyphonic performances.
-
4.
What is the difference in the model between harmonic dependency and melodic dependency?
-
5.
What dynamic programming technique can be efficiently used to calculate the musical expression?
-
6.
In what forum has Polyhymnia been independently evaluated, and what was the result?
-
7.
What statistical method with hidden state transition functionality is at the heart of Polyhymnia’s learning and modeling algorithm?
-
8.
Why is MusicXML able to encode much better than MIDI?
-
9.
In a crescendo or diminuendo, if sound intensity is changing exponentially, what is the human perception of the loudness?
-
10.
In Polyhymnia, how is loudness determined for mordents, turns, trills, and grace notes?
Rights and permissions
Copyright information
© 2013 Springer-Verlag London
About this chapter
Cite this chapter
Kim, T.H., Fukayama, S., Nishimoto, T., Sagayama, S. (2013). Statistical Approach to Automatic Expressive Rendition of Polyphonic Piano Music. In: Kirke, A., Miranda, E. (eds) Guide to Computing for Expressive Music Performance. Springer, London. https://doi.org/10.1007/978-1-4471-4123-5_6
Download citation
DOI: https://doi.org/10.1007/978-1-4471-4123-5_6
Published:
Publisher Name: Springer, London
Print ISBN: 978-1-4471-4122-8
Online ISBN: 978-1-4471-4123-5
eBook Packages: Computer ScienceComputer Science (R0)