Abstract:

This paper presents a prototype allowing the control of a concatenative synthesis algorithm using a 2D sketching interface. The design of the system is underpinned by a preliminary discussion in which isomorphisms between auditory and visual phenomena are identified. We support that certain qualities of sound and graphics are inherently cross-modal. Following this reasoning, a mapping strategy between low-level auditory and visual features was developed. The mapping enables the selection of audio units based on five feature data streams that derive from the statistical analysis of the sketch.