Real-time spatial processing of sounds for music, multimedia and interactive human-computer interfaces

Abstract

This paper gives an overview of the principlesand methods for synthesizing complex 3D sound scenesby processing multiple individual source signals. Signal-processing techniques for directional sound encoding andrendering over loudspeakers or headphones are reviewed,as well as algorithms and interface models for synthesizingand dynamically controling room reverberation and distanceeffects. A real-time modular spatial-sound-processing software system, called Spat, is presented. It allows reproducingand controling the localization of sound sources in three dimensions and the reverberation of sounds in an existing orvirtual space. A particular aim of the Spatialisateur project isto provide direct and computationally efficient control overperceptually relevant parameters describing the interactionof each sound source with the virtual space, irrespective ofthe chosen reproduction format over loudspeakers or headphones. The advantages of this approach are illustrated inpractical contexts, including professional audio, computermusic, multimodal immersive simulation systems, and architectural acoustics.