RESEARCH ARTICLE Predictive Models for Music

to achieve with traditional machine learning methods. This problem occurs when considering music data. In this paper, we introduce predictive models for melodies. We decompose melodic modeling into two subtasks. We first propose a rhythm model based on the distributions of distances between subsequences. Then, we define a generative model for melodies given chords and rhythms based on modeling sequences of Narmour features. The rhythm model consistently outperforms a standard Hidden Markov Model in terms of conditional prediction accuracy on two dierent music databases. Using a similar evaluation procedure, the proposed melodic model consistently outperforms an Input/Output Hidden Markov Model. Furthermore, these models are able to generate realistic melodies given appropriate musical contexts.

[1]  Simon Dixon,et al.  Evaluation of the Audio Beat Tracking System BeatRoot , 2007 .

[2]  François Pachet,et al.  The Continuator: Musical Interaction With Style , 2003, ICMC.

[3]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[4]  S. Handel,et al.  Listening: An Introduction to the Perception of Auditory Events , 1993 .

[5]  David Barber,et al.  A graphical model for chord progressions embedded in a psychoacoustic space , 2005, ICML.

[6]  E. Narmour.  The Analysis and Cognition of Melodic Complexity: The Implication-Realization Model , 1992 .

[7]  Douglas L. T. Rohde,et al.  Methods for Binary Multidimensional Scaling , 2002, Neural Computation.

[8]  Malcolm Slaney,et al.  A Unified System for Chord Transcription and Key Extraction Using Hidden Markov Models , 2007, ISMIR.

[9]  C. D. Gelatt,et al.  Optimization by Simulated Annealing , 1983, Science.

[10]  Ramon López de Mántaras,et al.  Melody retrieval using the Implication/Realization Model , 2005 .

[11]  David G. Stork,et al.  Pattern Classification , 1973 .

[12]  Jean-François Paiement,et al.  Probabilistic Melodic Harmonization , 2006, Canadian Conference on AI.

[13]  Jeff A. Bilmes,et al.  A gentle tutorial of the em algorithm and its application to parameter estimation for Gaussian mixture and hidden Markov models , 1998 .

[14]  Eric D. Scheirer,et al.  Tempo and beat analysis of acoustic musical signals. , 1998, The Journal of the Acoustical Society of America.

[15]  E. Schellenberg,et al.  Simplifying the Implication-Realization Model of Melodic Expectancy , 1997 .

[16]  A. F. Smith,et al.  Statistical analysis of finite mixture distributions , 1986 .

[17]  Christopher K. I. Williams,et al.  Harmonising Chorales by Probabilistic Inference , 2004, NIPS.

[18]  S. Handel.  Listening As Introduction to the Perception of Auditory Events , 1989 .

[19]  Eugene Narmour,et al.  The Analysis and Cognition of Basic Melodic Structures: The Implication-Realization Model , 1990 .

[20]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[21]  Jürgen Schmidhuber,et al.  Finding temporal structure in music: blues improvisation with LSTM recurrent networks , 2002, Proceedings of the 12th IEEE Workshop on Neural Networks for Signal Processing.

[22]  Yoshua Bengio,et al.  Learning long-term dependencies with gradient descent is difficult , 1994, IEEE Trans. Neural Networks.

[23]  Yoshua Bengio,et al.  Input-output HMMs for sequence processing , 1996, IEEE Trans. Neural Networks.

[24]  Shlomo Dubnov,et al.  Using Machine-Learning Methods for Musical Style Modeling , 2003, Computer.