Music Composition Using Harmony Search Algorithm

摘要

Music pieces have been composed using a behavior-inspired evolutionary algorithm, harmony search (HS). The HS algorithm mimics behaviors of music players in an improvisation process, where each player produces a pitch based on one of three operations (random selection, memory consideration, and pitch adjustment) in order to find a better state of harmony which can be translated into a solution vector in the optimization process. When HS was applied to the organum (an early form of polyphonic music) composition, it could successfully compose harmony lines based on original Gregorian chant lines.