Permeable Interdisciplinarity: Algorithmic Composition, Subverted
Aaron Einbond, City University, London
Artemi Maria Gioti, University College London
This project centres on reflexive auto-ethnographies of interdisciplinarity on the part of two composer-researchers (Aaron Einbond and Artemi-Maria Gioti) engaging with AI as a creative and collaborative tool. It asks: how can algorithmic compositional methods utilising AI be subverted creatively, and what is the role of interdisciplinary collaboration in this process? Questions of material engagement, musical labour, distributed creativity, and subjectification as they relate to AI are investigated in two interrelated studies exploring a range of interdisciplinary modes from ‘agonistic-antagonistic’ interactions between composer-researchers to ‘interdisciplinarity in one person’ (Barry and Born 2013).
The project is centred around the production of two compositions for acoustic instruments and electronics incorporating Machine Learning (ML): Aaron Einbond’s Prestidigitation, for percussion and 3-D electronics, and Artemi-Maria Gioti’s Bias II, for piano and interactive music system. Engaging with STS methodologies, the two studies go ‘inside’ the ML algorithms to explore how they ‘learn’ from Music Information Retrieval (MIR) data and human interpretative choices to challenge traditional notions of musical authorship and reshape the relationships between composers, performers, developers and listening subjects.
The project is based at UCL, with connections to and residencies at IRCAM (Institut de Recherche et Coordination Acoustique/Musique) in Paris (Einbond) and ZKM (Zentrum für Kunst und Medien) in Karlsruhe (Gioti), the latter via a ZKM commission following receipt of a Giga-Hertz production award. Both pieces result from collaborations with the performers: with percussionist Maxime Echardour (Einbond) and pianists Magda Mayas and Xenia Pestova Bennett (Gioti).