Real-time Algorithmic Timbral Spatialisation: Compositional Approaches and Techniques

This paper introduces a series of techniques and approaches to algorithmic timbral spatialisation, the realtime processing of audio data and its musical organization in a binaural or multichannel listening space. The intent of the work is to explore different ways for the automated spatialisation of the audio spectrum, especially in contexts of electroacoustic and acousmatic music composition. Typical Digital Signal Processing operations and algorithms will be used in order to create and/or retrieve data for the dynamic positioning of the audio sources. Firstly, in order to describe the processes, the concept of timbral spatialisation will be introduced, describing the compositional interest of such approach. Then the different techniques for data generation will be formalized, describing their “audioparous” process, that is when information regarding musical organization is extrapolated from an audio source. The various approaches to interpretation and usage of the data will be discussed, as well as their implementation in SuperCollider