Spatializing Timbre With Corpus-Based Concatenative Synthesis

Corpus-based concatenative synthesis presents unique possibilities for the visualization of audio descriptor data. These visualization tools can be applied to sound diffusion in the physical space of the concert hall using current spatialization technologies. Using CATART and the FTM&CO library for MAX/MSP we develop a technique for the organization of a navigation space for synthesis based on user-defined spatial zones and informed by the perceptual concept of timbre space. Spatialization responds automatically to descriptor data and is controllable in real-time or can be recorded for later playback. This work has been realized in recent compositions for instruments, electronics, and sound installation using Wave Field Synthesis and Vector-Based Amplitude Panning. The goal is to place the listener in the midst of a virtual space of sounds organized by their descriptor data, simulating an immersion in timbre space.