VA Mixing Implementation in Virtual Acoustic Space

This session aims to further develop understanding of the interactions between receiver positioning relative to audio sources within VAS. For this session, an arrangement of string instruments will be positioned in a theatre model to synthesise the listening experience of orchestral music. Whilst these sources will remain static, the listening position will be moved for the purpose of testing microphone placement and pickup within the VAS.

The multitracks in use are of La Lyra - Telemann performed by KCL’s Baroque orchestra - and contain stems from a decca tree microphone placement. This trio of microphones captures the whole orchestra from a height, and therefore source placements within the VAS will correspond to this. There are also three additional spot sources of violin / viola / cello.

The varying positions of the receiver in the VAS changes spectral characteristics of depth to the final render, specifically those attributed to distance including reflections / attenuation / absorption / transmission. The further from a soundsource the receiver was, the more diffuse the sound with localisation becoming wider around the listener. The side positioned recordings give some more insight on the propagation of audio taking place. With the left receiver recording, the sound is diffused and ambient in such a way that the specific directionality is masked as the sound fills the VAS, as direct sound is being layered between many reflection cues coming in from all directions.

With the left being positioned nearer centre then the right, there is a clear distinction between the two based on distance on the horizontal axis of positioning. There is also occlusion taking place in the right side recording, as directly in front of the receiver in the entrance there is the right side wall of the theatre model. This occlusion results in a void space within the stereo field where incoming audio and reflection cues do not cast through, this results in a heightened definition between incoming audio and space within the mix for other components. This could prove to be beneficial for layering VAS sounds with other audio tracks, where stereo zones can be occluded in render.