The Well Trained Algorithm is a composition that challenges prevailing conceptions of the use of AI tools in music through the reconceptualising of JukeBox, a generative AI model for music, as an instrument in its own right. Here, I am coining the term ‘instrumentisation’ to describe a methodology for applying the qualities and associations of a musical instrument to a traditionally non-musical object. To showcase this conceptual approach, this model of thinking is applied to aid in the composition of the AI-generated musical piece, The Well Trained Algorithm. Through this process of ‘instrumentisation’, musical terms such as tuning and timbre are redefined to better relate to the specific affordances of an artificially intelligent system. The composition is informed, then, by an exploration of a system's instrumental possibilities, leading to a more effective and artistic use of the technology in the creative process. The seminal works of J. S. Bach and La Monte Young, The Well Tempered Clavier and The Well Tuned Piano, respectively, provide a historical, musical, and theoretical context for the piece as well as the datasets used to fine-tune the JukeBox model. With this thesis, I ask if, through a process of ‘instrumentisation’ AI technology can be successfully reconceptualised as a musical instrument as a means to promote artistic expression.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:kmh-4998 |
| Date | January 2023 |
| Creators | O'Riain, Muiredach |
| Publisher | Kungl. Musikhögskolan, Institutionen för komposition, dirigering och musikteori |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0027 seconds