A methodology for the comparison of melodic generation models

We investigate musical metacreation algorithms by applying Music Information Retrieval techniques for comparing the output of three off-line, corpus-based style imitation models; Variable Order Markov Chains, the Factor Oracle, and MusiCOG. Our focus is on discovering which musical biases are introduced by the models, that is, the characteristic of the output which are shaped directly by the formalism of the models and not by corpus itself.


Nicolas Gonzalez ThomasPhilippe PasquierArne Eigenfeldt, James Maxwell

Research paper and Posters

Gonzalez Thomas, N., Pasquier, P. & Eigenfeldt, A. & Maxwell, J. B. (2013). “A METHODOLOGY FOR THE COMPARISON OF MELODIC GENERATION MODELS USING META-MELO” Proceedings of the 14th International Society for Music Information Retrieval Conference (ISMIR 2013), Curitiba, Brazil, 561-566.

Download PDF