Using Machines to Define Musical Creativity
Abstract
This dissertation aims to assess whether machines are capable of creativity in music composition, how they are capable of musical creativity and whether this helps to define what musical creativity is. It will also discuss how these machine systems can be useful for our own creative processes. This dissertation will focus on two case studies: the hacker-duo Dadabots who generate raw audio using SampleRNN, and tech start-up Auxuman who create synthetic humans who are capable of generating music and lyrics through artificial neural networks. Through evaluating these case studies, it will be argued that current definitions of creativity in music composition are not broad enough.
The findings from evaluating Dadabots’ work will firstly demonstrate that our assumptions about musical creativity in both humans and machines revolve too strongly around symbolic models. Secondly, the findings will suggest that what Boden describes as ‘transformational creativity’ can take place through unexpected machine consequences. (1)
The chapter on Auxuman will focus on the significance of musical meaning to the creative process and through assessing Auxuman’s output will argue that meaning is codable, can be an appropriate goal when composing (or generating) music and thus, can be a significant part of musical creativity.
(1) Margaret Boden, The Creative Mind: Myths and Mechanisms (London: Routledge, 2004), p. 3