Computer Systems Research Laboratory
Bell
Laboratories --- Lucent Technologies, Room 2C304
600
Mountain Avenue, Murray Hill, NJ 07974 USA
wolf@research.bell-labs.com, Tel: (++1) (908) 582-8011
In this paper, we will discuss, from a general perspective, the design and implementation of a system called WOLFGANG that composes tonal monodies. Our examination will focus on defining the evaluation criteria guiding WOLFGANG's compositional processing. The thesis of the work presented is derived from the hypothesis that a system's ``innate'' sense of musical sound strongly influences the development of its ``perception'' as well as composing habits. As the system develops its musical skills, it also develops a subjective use of a musical language biased by its sense of musical sound, and its adaptatio n to the cultural grammar of its environment.
In essence, a person becomes a composer via many diverse experiences originating from emmercement in a cultural. Further, a composer continues to learn and modify composing techniques as a direct reflection of changes to the current culture. Thus, it is necessary t o design systems which represent and apply knowledge learned from cultural experiences. Systems such as WOLFGANG, must be designed with extensive and flexible knowledge representations so as to support (1) strong plan reformulation functions and (2) strong abstraction capability. It is plausible to expect such systems (3) to be interactive assista nts in a broad range of tasks, not just limited to composing.
Keywords: spreading-activation, K-lines, dispositions, emotions.