Solving the musical orchestration problem using multiobjective constrained optimization with a genetic local search approach

Abstract

In this paper a computational approach of musical orchestration is presented. We consider orchestration as the search of relevant sound combinations within large instruments sample databases and propose two cooperating metaheuristics to solve this problem. Orchestration is seen here as a particular case of finding optimal constrained multisets on a large ensemble with respect to several objectives. We suggest a generic and easily extendible formalization of orchestration as a constrained multiobjective search towards a target timbre, in which several perceptual dimensions are jointly optimized. We introduce Orchidée, a time-efficient evolutionary orchestration algorithm that allows the discovery of optimal solutions and favors the exploration of non-intuitive sound mixtures. We also define a formal framework for global constraints specification and introduce the innovative CDCSolver repair metaheuristic, thanks to which the search is led towards regions fulfilling a set of musical-related requirements. Evaluation of our approach on a wide set of real orchestration problems is also provided.

Constraint Programming Approach to the Problem of Generating Milton Babbitt’s All-Partition Arrays

“…4 For a more detailed description of the constraints governing the organization of matrices in Babbitt's music, see [2,8]. 5 We denote an integer partition of an integer, L, by IntPartL(s1, s2, . .…”

“…Using our model and a method for dividing this matrix into smaller, sub-problems, we obtained a solution, which, we believe, is the first to be discovered automatically using a computer and differs from those found by composers. CP is a programming paradigm that has been successfully applied to the solving of various constraint satisfaction problems in music [3][4][5][6][7]. It seems natural then, that CP could be used in the problem we address here.…”

Sample Orchestrator : gestion par le contenu d’échantillons sonores

“…Le phénotype, quant à lui, correspond aux propriétés perceptives de chaque individu, qui s'expriment dans un espace métrisé, (donc dans lequel on mesure des distances). En pratique, l'algorithme retenu s'inspire du MOGLS (Jaszkiewicz, 2001) : l'évaluation des individus se fait à l'aide de normes de Tchebycheff pondérées, desquelles on peut explicitement déduire un jeu de préférences subjectives quantifiées, et utiliser cette information dans les étapes ultérieures du calcul (Carpentier et al, 2010b). …”

Dynamic Musical Orchestration Using Genetic Algorithms and a Spectro-Temporal Description of Musical Instruments

“…Secondly, a client/server architecture is put forward, providing the users with the opportunity of implementing their own control interfaces while keeping the computational issues in the core program. Last, Orchidée explicitly addresses the timbre multidimensionality and the combinatorial complexity of instrument mixtures by the joint use of a multicriterion optimization and evolutionary algorithms [2]. Orchidée has already been used by many notable composers, among them Jonathan Harvey for his latest piece Speakings [8].…”