We have several related interests in this area that range from finding new abstract representations of music (notation) that better enable analysis and understanding of musical structure to ways of compactly encoding musical sounds that best retain the essential elements of musical expression. In the former area we are exploring mappings of musical structure onto geometric spaces to gain insights into the unfolding of musical structure, and the application of principles of complexity theory to a “holistic” model of musical design. We have the ambitious goal of developing a dynamical theory of musical structure evolution in analogy to the variational (or minimal) principles that underlie many physical theories.
Another endeavor in this area has been to employ physical modeling of musical instruments as a means to encode musical sounds. Specifically we use the time-history of the control parameters of appropriate physical models to capture and represent the essential expressive features of musical sound. This new encoding strategy promises to have the compactness of a MIDI and the expressiveness of a high quality digital recording and may form the basis of a new, very compact musical language that can form the basis of the interaction between musician and electronic sound synthesizers.