Music Composition with Interactive Evolutionary Computation

Abstract

Interactive Evolutionary Computation (IEC), i.e., Evolutionary Computation whose fitness function is provided by a user his/herself, has been applied to esthetic areas, such as art, design and music. We cannot necessarily define fitness functions explicitly in these areas. With IEC, however, we can embed the user's implicit preference into the optimization system. This paper describes a new approach to music composition, more precisely the composition of rhythms, by means of IEC. The main feature of our method is to combine Genetic Algorithms (GA) and Genetic Programming (GP). In our system, GA individuals represent short pieces of rhythmic patterns, while GP individuals express how these patterns are arranged in terms of their functions. Both populations are evolved interactively through the user's evaluation. The integration of interactive GA and GP makes it possible to search for musical structures effectively in the vast search space. In this paper, we show how successfully our proposed method can generate attractive musical rhythms. The effectiveness of our system is demonstrated by the evolved rhythm phrases, which are available from our web site as sound files.