Perceptual grouping of musical sounds: A computational model

There is substantial recent evidence that the mixture of sounds reaching the ears is subjected to an analysis by early auditory processes, in which acoustic components with common properties are grouped into a single percept. Since most music consists of a mixture of sounds, it might be expected that early auditory processing also mediates the perception of musical sounds. In this article, we present a computational model of some aspects of auditory processing, and investigate the grouping of musical sounds using the model. In particular, a scheme is developed for grouping sounds across time according to their timbres, as represented within a two‐dimensional space of “brightness” and onset asynchrony.