Melodizer 2.0 : a constraint programming tool for computer-aided musical composition

Details

Supervisors

Van Roy, Peter

Faculty

Ecole polytechnique de Louvain
Ecole polytechnique de Louvain

Degree label

Master [120] : ingénieur civil en informatique, à finalité spécialisée
Master [120] : ingénieur civil en mathématiques appliquées, à finalité spécialisée

Abstract

This master’s thesis presents the design of a tool destined to assist musical composers in the creation of their next masterpiece. The composers state the musical ideas that they want to include in their themes which are translated into a Constraint Satisfaction Problem. This thesis develops two already ex- isting master’s thesis. The first one, written by Baptiste Lapière, was more rhythm-oriented [10]. While the second one, written by Damien Sprockeels, was more focused on pitch-oriented scenarios [28]. Therefore, we combined both works to create a tool that allows to play with pitches and rhythms simultaneously. On the one hand, Gecode is a powerful C++ toolkit that is used in order to model and solve Constraint Optimization Problems. While, on the other hand; OpenMusic, based on Lisp, serves as the visual programming and composition environment where Melodizer 2.0 is employed. GiL works as the bridge between Gecode and Lisp that allows us to solve Constraint Satisfaction Problems in Openmusic. Melodizer 2.0 provides an intuitive interactive interface that works as a melody synthesizer with many knobs and buttons to tweak in search of inspiring results. We do not pretend to replace musician’s creativity nor come up with a full master- piece when launched. Nevertheless, it stimulates songwriters in their production process. If you are an inspired compositor that is eager to use Melodizer 2.0 we recommend you to go directly to chapters 6 and 7 where we explain how to manipulate the interface, and, provide plenty of musical scenarios to picture the different uses and the musical relevance of our instrument.