Virtual Acoustic Space Models
The following VAS model rooms have been created to generate varying audio rendering environments which the signal stems can propagate, encoding them with the virtual acoustic properties of each. These examples use different stems for each source, and vary in VAS sizes to investigate the breadth of variety which could be possible with this encoding process. Movement of sound sources are present in these recordings, and has been done to demonstrate the localisation properties of the HRTF as well as highlight the variation in sound’s timbre / spectra from reflections dependent on listening position. VAS Room models have been created using SketchUp software, scaled to the average height model and sequentially increasing in their volume.
It is worth noting that the arrangement approach to layering sounds in the mix through movement around the space is a potentially lucrative approach to spatial mixing where audio sources are not fixed in position relative to the receiver. The multichannel VAS model render highlights this, as the movement in and out of the near surround field in tandem with the relative movement of incoming reflections makes for an immersive and interesting depth to the final recording, and a creative exploratory approach to multitrack layering.
Positioning greatly impacts the encoded properties, and exploring around the space to find these spots within the VAS leads to the discovery of unitindented timbre with positive potential application. As with the long hall model, reflections present a very distinct ITD from the far end reflections when the HRTF receiver is positioned centrally, which makes for interesting rhythmic delay patterns based on location / VAS reflection cues.
Finally, the short clipped amplitude points within the audio result from low-end level increase when nearing walls or corners too closely, simulating the acoustic property of standing waves. These standing waves and VAS rooms modes could be an area to investigate to understand signal propagation of virtual acoustics further. Below are demonstrations of early stage Implementations prior to the in depth iteration development of VA and VAS encoding documented herein.
Figure 11 : Small Club Model - Electronic Track Audio Source (Sharp, 2023) VA TestSession - Club VAS.wav https://drive.google.com/file/d/19lVTT3lATmMWCQbS14ztmrqCIfOYpHDD/view?usp=share_link
Figure 12 : Small / Medium Room Model [club model left for scale], Source For Strings / Piano (Sharp, 2023) VA TestSession - Strings / Piano Small Medium Size VAS.wav https://drive.google.com/file/d/13SqtxtGE0I7GFMDXfnJn8QKk6BfbqaVa/view?usp=share_link
Figure 13 : Medium Sized Room Model With Additional Height, Multi-Channel Arrangement Of 5 Sources Of Classical Instrumentation (Sharp, 2023)
VA TestSession - MultiChannel Medium Size VAS.wav https://drive.google.com/file/d/1TS_LI2ie3niHkE7raXwLGb30-klzUqtu/view?usp=share_link
Figure 14 : Long Hall Model, Single String Audio Source At The Far End From The Door (Sharp, 2023) VA TestSession - Plucked Strings Long Hall VAS.wav https://drive.google.com/file/d/1k9C2pOrL_dIt6yXBjWdAdrBFABQQLs0E/view?usp=share_link
Figure 15 : Large Tall Room Model With Domed Ceiling, Two Audio Sources For Violin / Viola Stems At The Base Of The Room (Sharp, 2023)
VA TestSession - Violins Tall Domed Room VAS.wav https://drive.google.com/file/d/1xSichgxT3Xn6UbgRYZ9TjLVAItJNPExO/view?usp=share_link
Figure 16 : Short diffusive model with horn / onion shaped dome, Single flute positioned in the ceiling space (Sharp, 2023)
VA TestSession - Flute Horn Domed VAS.wav https://drive.google.com/file/d/1Q-ILrLcg2rJhD3uhnakiyxvpN5RHlyQp/view?usp=share_link
