Generative Sonic Voyage,
|Performance presentation for GA2003

Chun Lee, BA (Hons), MA

Lansdown Centre forElectronic Arts, Middlesex University, London, England.

e-mail: leechun@leechun.freeserve.co.uk

Abstract

Genetic algorithms have great potential to be applied as compositional methods in generative music. This is predominantly because the process generated by genetic algorithms is accumulative, which somewhat reflects the conventional notion of musical structure resulting from a process of growth and development. Genetic algorithms, when applied to certain musical parameters, are particularly suitable for creating structures in composition. Moreover, the mechanisms in genetic algorithms such as crossover and mutation, can, through recombination, give rise to novel results within the chosen musical elements. Lastly, because genetic algorithms explore different evolutionary paths when presented with the same problem, this distinctive feature can be applied to generate music that has recognisable characteristics with ever changing details.

The presentation demonstrates the use of programming techniques derived from genetic algorithms applied to live computer music performance. It also displays the diversity of musical styles that can be achieved with the use of genetic algorithms. These include abstract sonic-sculpture, real-time feedback signal manipulation and contemporary beat orientated electronica music.

The performance can last from 30 to 60 minutes. The creative system used in the live performance is an interactive compositional environment programmed in Max/Msp in conjunction with custom made external genetic algorithm objects created by the performer and used by him for over two years. Max/Msp is an object oriented GUI programming environment widely used in creating interactive and mixed media works.