Cellular automata sound synthesis: from histograms to spectrograms.
Jaime Serquera, Eduardo Reck Miranda · 2008
Abstract. We are investigating ways in which advanced unconventional computation methods may provide novel approaches to music technology. In this paper we report on a new technique to synthesise sounds from Cellular Automata (CA) models of reaction-diffusion chemical computers. We render the behaviour of CA into sounds using statistical analysis of CA cells as their values change in time. We estimate the probability distributions of the cells values for each cycle of the automaton by histogram measurements of the images produced by plotting the values of the CA cells as a matrix of coloured pixels. We have studied the behaviour of a CA model proposed by Gerhard and Schuster under a variety of different settings in order to gain a better understanding of the histograms, with a view on predicting the types of sounds that different CA behaviours would produce. 1