TitleMusic Performance Generation as Time Series Prediction
Publication TypeConference Paper
Year of Publication2004
AuthorsGrachten M, Arcos JLluis
EditorGervás P., Gupta K.M
Conference NameProceedings of the ECCBR 2004 Workshops
PublisherDepartamento de Sistemas Informáticos y Programación. Universidad Complutense de Madrid
Pagination329-336
Abstract

In this paper, we present a case-based reasoning system for rendering expressive music performances in the context of popular jazz themes. Music performance by skilled musicians is far from being a literal translation process from notes written at the score to sounds with the nominally described attributes. These score deviations can vary greatly from one musician to another musician and represent the musician style. The case-based reasoning system presented deals with this problem as a time series prediction problem.