An Evaluation of Score Descriptors Combined with Non-linear Models of Expressive Dynamics in Music

Expressive interpretation forms an important but complex aspect of music, in particular in certain forms of classical music. Modeling the relation between musical expression and structural aspects of the score being performed, is an ongoing line of research. Prior work has shown that some simple numerical descriptors of the score (capturing dynamics annotations and pitch) are effective for predicting expressive dynamics in classical piano performances. Nevertheless, the features have only been tested in a very simple linear regression model. In this work, we explore the potential of a non-linear model for predicting expressive dynamics. Using a set of descriptors that capture different types of structure in the musical score, we compare the predictive accuracies of linear and non-linear models. We show that, in addition to being (slightly) more accurate, non-linear models can better describe certain interactions between numerical descriptors than linear models.

[1]  Roberto Bresin,et al.  Artificial neural networks based models for automatic performance of musical scores , 1998 .

[2]  David P. Helmbold,et al.  Modeling, analyzing, and synthesizing expressive piano performance with graphical models , 2006, Machine Learning.

[3]  Gerhard Widmer,et al.  Computational Models of Expressive Music Performance: The State of the Art , 2004 .

[4]  D. Temperley.  Music and probability , 2006 .

[5]  Nell P. McAngusTodd,et al.  The dynamics of dynamics: A model of musical expression , 1992 .

[6]  A. Gabrielsson.  Music Performance Research at the Millennium , 2003 .

[7]  Gerhard Widmer,et al.  Machine Discoveries: A Few Simple, Robust Local Expression Principles , 2002 .

[8]  Gerhard Widmer,et al.  Discovering simple rules in complex data: A meta-learning algorithm and some surprising musical discoveries , 2003, Artif. Intell..

[9]  Nitish Srivastava,et al.  Improving neural networks by preventing co-adaptation of feature detectors , 2012, ArXiv.

[10]  Gerhard Widmer,et al.  The Magaloff Project: An Interim Report , 2010 .

[11]  Paola Annoni,et al.  Variance based sensitivity analysis of model output. Design and estimator for the total sensitivity index , 2010, Comput. Phys. Commun..

[12]  C. Palmer.  Music performance. , 1997, Annual review of psychology.

[13]  W Goebl,et al.  Melody lead in piano performance: expressive device or artifact? , 2001, The Journal of the Acoustical Society of America.

[14]  Freya Bailes,et al.  Performing Musical Dynamics: How Crucial are Musical Imagery and Auditory Feedback for Expert and Novice Musicians? , 2014 .

[15]  P. Juslin.  Communicating emotion in music performance: A review and a theoretical framework , 2001 .

[16]  Shin-ichi Maeda,et al.  Gaussian Process Regression for Rendering Music Performance , 2008 .

[17]  J. Sloboda.  Generative Processes in Music: The Psychology of Performance, Improvisation, and Composition , 1985 .

[18]  Gerhard Widmer,et al.  Analysis and prediction of expressive dynamics using Bayesian linear models , 2014 .

[19]  G. Widmer,et al.  BAYESIAN LINEAR BASIS MODELS WITH GAUSSIAN PRIORS FOR MUSICAL EXPRESSION , 2015 .

[20]  C. Palmer.  Anatomy of a Performance: Sources of Musical Expression , 1996 .

[21]  Harm de Vries,et al.  RMSProp and equilibrated adaptive learning rates for non-convex optimization. , 2015 .

[22]  J. Sundberg,et al.  Overview of the KTH rule system for musical performance. , 2006 .

[23]  Radford M. Neal.  Pattern Recognition and Machine Learning , 2007, Technometrics.

[24]  D. Shahsavani,et al.  Variance-based sensitivity analysis of model outputs using surrogate models , 2011, Environ. Model. Softw..

[25]  Rafael Ramirez,et al.  Rule induction for expressive music performance modeling , 2004 .

[26]  J. Sundberg.  Summary of the Music Performance Panel , MOSART Workshop 2001 , 2002 .

[27]  W. L. Windsor,et al.  Expressive Timing and Dynamics in Real and Artificial Musical Performances: Using an Algorithm as an Analytical Tool , 1997 .

[28]  Gerhard Widmer,et al.  Linear Basis Models for Prediction and Analysis of Musical Expression , 2012 .

[29]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[30]  Giovanni De Poli.  Analysis and modeling of expressive intentions in music performance. , 2003, Annals of the New York Academy of Sciences.

[31]  Alex Graves,et al.  Generating Sequences With Recurrent Neural Networks , 2013, ArXiv.

[32]  E. Clarke.  Generative principles in music performance. , 1988 .

[33]  J. Sloboda.  The Communication of Musical Metre in Piano Performance , 1983 .

[34]  Maarten Grachten,et al.  Predicting Expressive Dynamics in Piano Performances using Neural Networks , 2014, ISMIR.

[35]  Douglas Eck.  Beat Tracking using an Autocorrelation Phase Matrix , 2007, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07.