Creating and Exploring Huge Parameter Spaces: Interactive Evolution as a Tool for Sound Generation

Abstract

In this paper, a program is presented that applies\ninteractive evolution to sound generation, i.e., preferred\nindividuals are repeatedly selected from a population of\ngenetically bred sound objects, created with various\nsynthesis and pattern generation algorithms. This simplifies\naural exploration of huge synthesis parameter spaces, and\npresents a possibility for the sound artist to create new\nsound engines customized for this kind of creation and\nexploration – sound engines too complex to control in any\nother way. Different sound engines are presented, together\nwith a discussion of compositional applications. It is also\nshown how this technique can be used to simplify sound\ndesign in standard hardware synthesizers, a task normally\navoided by most musicians, due to the required amount of\ntechnical understanding.