Proposing Spatdif - the Spatial sound Description Interchange Format

This paper outlines the requirements for an interchange format that can describe and share spatial parameters across 3D audio applications, and proposes SpatDIF for its implementation. 1. WHY USING A SCENE DESCRIPTION FORMAT FOR COMPOSING? Formats as a structuring concept are integral to musical practice. For example, in the form of scores, a written symbolic representation of music, compositions can be stored, exchanged, studied, performed but also revised and adapted after their initial creation. MusicXML shows how the score concept is digitally maintained. Although spatialization can be considered as a core element of electroacoustic music, there is no general consensus in how to describe and notate spatialization. Nowadays the spatial aspects are mostly created and automatized on a low-level within diverse digital audio environments, such as Max/MSP or ProTools. Because these environments have different syntaxes, units, and storage solutions, the control messages (e.g. a trajectory to move a sound in space) are only valid within this specific audio environment. Therefore the interchangeability of these descriptors is ineffectual and usually spatial aspects are directly rendered into multichannel sound files. Now that processing power is usually sufficient to render multiple virtual sound sources in real-time, a separation of “raw” sound material from the spatial descriptors within an open data format would increase the portability across different 3D audio applications, loudspeaker configurations and concert venues. Furthermore, spatial rendering algorithms could be compared and combined without having to change the spatial-sound syntax. This, of course, relies on the standardization of descriptors.