DESCRIPTION-DRIVEN CONTEXT-SENSITIVE EFFECTS

Abstract

We introduce a new paradigm in digital audio effects that is based on more symbolic manipulations of elements of a sound, rather than using linear signal processing alone. By utilising content descriptions such as those enabled by MPEG-7, a system may apply context-sensitive effects that are more aware of the structure of the sound than current systems. We advocate a standards-based approach (with MPEG-4, - 7, and -21) so as to maximise the interoperability between different systems. The paper outlines MPEG-7 description structures that may be used as the basis for controlling and triggering effects in a system. It explores the different possibilities that are opened up by this paradigm. The way is then pointed towards more sophisticated control structures that may lead to more “musical” and dynamic effects.