Machine Listening as a Generative Model: Happy Valley Band

ORGANVM PERCEPTVS is a collection of 11 songs for mixed ensemble written by translating machine listening analysis of pop songs into musical notation. Motivated by the idea that analysis algorithms inherently carry the values of the communities that produce them, I see my compositional process as a way of understanding how musical ideas, beliefs, preferences, and aesthetics are embedded within analysis algorithms. The musical scores are overly-specific, complex, and brimming with the artifacts of the machine listening process, and I formed a dedicated ensemble, the Happy Valley Band, to develop a performance practice unique to the idiosyncratic music. This dissertation essay documents the analysis algorithms used, my compositional process, the performance practice developed, the process of recording and releasing an album of music, and the public reception. I discuss my compositional ideas in the context of a number of twentieth century aesthetic traditions, including plunderphonics and sampling, computer analysis driven composition, and complexity in computer-assisted composition. I see this project as relevant to emerging cultural concerns of algorithmic bias and discrimination, and I relate my experience to contemporary dialogues around digital automation.