Published online by Cambridge University Press: 11 January 2023
This article discusses Nautilus (2022), a composition for solo bass flute created using machine-learning techniques and a Unity game engine. We consider the approaches we adopted and how they enhanced creativity and musicianship for those involved. We reflect on Unity's potential as a novel and flexible driver for the creation of a musical score in which traditional elements of compositional design are presented to a performer as a co-creator for interpretation and communication inside the act of musicking. Through this we offer insights into performer agency and how a performer decodes media, sound, images and AI through their instrument, personal skills and musical aesthetic. We describe how the notion of a music score was re-conceptualised, transforming our understanding of the activities of composition, collaboration and performance.
1 Details are available online at https://cordis.europa.eu/project/id/101002086 (accessed 25 May 2022).
2 Small, Christopher, Musicking (Middleton: Wesleyan University Press, 1989)Google Scholar.
3 Ibid., p. 9.
4 Ibid. p. 13.
5 Emmerson, Simon, Living Electronic Music (London: Routledge, 2007), p. 29Google Scholar.
6 A neural network is a complex statistical analyser used to output useful information from a given set of input data – for example, to predict the house prices of a particular suburb based on previous sale, size, demographic, socio-economic and location data. The training code for Nautilus used this repository, with some minor changes: https://github.com/haryoa/note_music_generator/blob/master/Music%20Generator.ipynb (accessed 28 November 2022).
7 This is discussed in more detail at https://digiscore.dmu.ac.uk/2022/01/17/the-digital-score-through-the-medium-and-its-message/ (accessed 25 May 2022).
8 Mac and Windows versions of the digital score can be downloaded at https://digiscore.dmu.ac.uk/2022/01/27/nautilus/ (accessed 22 November 2022).