The Impact of the Digital Instrument and the Score on Controlled Improvisation When using Acoustic Instruments in an Electroacoustic Context

This paper considers various elements that influence the decisions taken when designing performance environments for compositions featuring acoustic instruments with electronics. The implications of choosing different models for the control of the electronics are discussed. Models lying along a continuum between fully automated systems at one end and passive environments controlled by a human performer at the other are considered. The ways in which choices made when mapping control parameters to the sound engine affect the affordances available to the human performer are explored, and the advantages of using a score to define parameters for improvisation are discussed. Following from this, different models for notating the electronic part in mixed compositions are presented and the implication of using the computer as the principal agent for controlling the electronic part are considered. This leads into a discussion as to what constitutes a score for the electronic part. Different paradigms are presented and their implications considered. The paper then discusses the ease of encoding and decoding musical information at different representational levels into and out of a computer. This has implications for machine learning and score following systems. The paper concludes by presenting an example of a digital performance environment from the author’s own work, together with a score for controlled improvisation using the interface.