STRUCTURING MUSIC BY MEANS OF AUDIO CLUSTERING AND GRAPH SEARCH ALGORITHMS

Frédéric Le Bel (1, 2)

1 IRCAM - Institut de Recherche et Coordination Acoustique/Musique
2 CICM - Centre de recherche Informatique et Création Musicale

This article proposes to explore a work in progress that aims at developing a computer-aided-composition (CAC) approach to structuring music by means of audio cluster- ing and graph search algorithms. Although parts of this idea have been investigated in order to achieve different tasks such as corpus-based concatenative synthesis, musical genre recognition or computer-aided orchestration to name a few, the challenge remains to find a way of integrating these techniques into the creative pro- cess; not to generate material but to explore, to analyse and to understand the full potential of a given sound cor- pus prior to scoring a musical piece. As opposed to main- stream CAC tools, mostly focusing on generative methods, the following one proposes an analytical approach to structuring music through the creative process. As the title of this article reveals some pieces of answer to this interrogation, the following topics aim at unfolding the algorithmic structure in order to examine the methodology, to discuss a few important co-lateral problematics, to expose the limitations of such an approach and finally to discuss ideas for future development.

Mots clés: graph search algorithms; audio clustering; audio features; musical structures; computer music; music informatics

Domaines: Musique, musicologie et arts de la scène

Frédéric Le Bel. STRUCTURING MUSIC BY MEANS OF AUDIO CLUSTERING AND GRAPH SEARCH ALGORITHMS. Journées d'informatique musicale 2017, Couprie P., Davy-Rigaux C., Genevois H., Liao L.-N., Malt M., Maniguet T., Mifune M.-F., Journées d'informatique musicale, Paris, Collegium Musicæ, 2017., May 2017, Paris, France. ⟨hal-01577224⟩