Music and Neural Networks

Abstract

This chapter pertains to the research in the field of music and artificial neural networks. The first attempts to codify human musical cognition through artificial neural networks are taken into account as well as recent and more complex techniques that allow computers to learn and recognize musical styles, genres, or even to compose music. Special topics covered are related to the representation of musical language and to the different systems used for solving them, from classic backpropagation networks to self-organizing maps and modular networks. The author hopes that this chapter will disclose some significant information about this emerging but nonetheless important subfield of AI and at the same time increase some interest and allow for a better understanding of this complex field.