Sound Speed Perturbation Robust Audio

Sankha Subhra Bhattacharjee, Jesper Rindom Jensen, Mads Græsbøll Christensen

Abstract: Sound zone control (SZC) algorithms rely on pre-recorded, static impulse responses (IRs) to construct the control filters. Hence, their performance is compromised by perturbations within the acoustic environment. An important perturbation is changes in sound speed, which result primarily from temperature and humidity variations. In this paper, we first examine the impact of sound speed variations on IRs, and both empirically and theoretically show that changes in speed (either increase or decrease) leads to a corresponding compression or expansion of the IRs along the time axis. We then introduce a simple model, termed sinc interpolation-compression/expansion-resampling (SICER), to correct the IRs for sound speed change. Our simulation studies demonstrate that the corrected IRs closely match the actual IRs, for both increase and decrease in sound speed. Furthermore, we discuss several important considerations while correcting IRs for sound speed variations. In the context of sound zone control, using the proposed SICER model, IRs pre-measured at a certain sound speed can be corrected for any sound speed change and optimal control filters at the new sound speed can be re-computed, without the need of re-measuring the new IRs (which becomes impractical after SZC deployment). We integrate the proposed SICER IR correction method with the recently proposed variable span trade-off (VAST) framework for SZC, and propose a SICER-corrected VAST method for sound zone control, resilient to sound speed variations. Simulation studies carried out show that the proposed SICER-corrected VAST method significantly improves acoustic contrast and reduces signal distortion in the presence of sound speed changes.

External IDs:doi:10.1109/taslpro.2025.3570949