eō: an evolutionary sound installation
This sound installation took place from the 13th-15th of October 2021 and was funded by the Berkofsky Arts Award.
The installation used a genetic algorithm to evolve microtonal vocal music in real-time. Though genetic algorithms are commonly used for optimisation purposes, the installation did not strive for perfection. Instead, it generated a ‘population’ of musical phrases and evolved them over time, through multiple generations, towards a desired ‘fitness’ (which was customisable). Throughout this process, musical phrases were born, reproduced, mutated, died out and were replaced, which allowed for both development of existing material and innovation. When running the algorithm, there was no way of telling whether the evolution would destroy the initial material, progress into something entirely new, result in a homogenisation of the material as it strived towards the fitness function, or some combination of these.
I used a non-standard tuning system for the installation in order to provide an alternative and focused listening experience, avoiding the musical evolution being perceived in terms of familiar melodic and harmonic concepts. I hoped that this would remove the listener’s attachment to preconceived expectations so that they could simply listen.
I have not included a recording of the installation audio because it was important for me that this piece was experienced in real-time as the algorithm evolved.