Generative Music with Stochastic Diffusion Search

This paper introduces an approach for using a swarm intelligence algorithm, Stochastic Diffusion Search (SDS) – inspired by one species of ants, Leptothorax acervorum – in order to generate music from plain text. In this approach , SDS is adapted in such a way to vocalise the agents, to hear their “chit-chat” . While the generated music depends on the input text, the algorithm’s search capability in locating the words in the input text is reflected in the duration and dynamic of the resulting musical notes. In other words, the generated music depends on the behaviour of the algorithm and the communication between its agents. This novel approach, while staying loyal to the original input text, when run each time, ‘vocalises’ the input text in varying ‘flavours’.