A round robin on room acoustical simulation and auralization.

Area of Science:

• Acoustics
• Computational physics
• Psychoacoustics

Background:

• Room acoustic modeling software is crucial for predicting sound behavior in spaces.
• Evaluating the accuracy of these models in both physical and perceptual aspects is essential for advancement.

Purpose of the Study:

• To assess the current state-of-the-art in room acoustic modeling software.
• To identify limitations in both physical prediction and perceptual realism of simulated acoustics.

Main Methods:

• A round-robin test involving multiple acoustic scenes and environments.
• Comparison of simulated results against measured data in physical and perceptual domains.

Main Results:

• Most simulation algorithms fail when geometrical acoustics assumptions are unmet, leading to errors in early reflection patterns and medium-to-high frequency predictions.
• Simulated auralizations were plausible but not authentic, with audible differences from measurements, particularly in tone color and source localization.

Conclusions:

• Current room acoustic models exhibit significant errors outside the scope of geometrical acoustics.
• Simplified modeling of absorption, scattering, and early reflections, including diffraction, limits perceptual authenticity.