Diffraction perception in L-shaped rooms using virtual reality

TL;DR

This paper explores how sound diffraction in L-shaped rooms affects auditory perception using virtual reality.

Contribution

It introduces and evaluates efficient IIR filter diffraction models for plausibility in acoustic simulations.

Findings

Including diffraction increased the perceived plausibility of acoustic simulations.

Diffracted reflection paths significantly improved plausibility in the shadow zone.

IIR filter models were similarly plausible to BTMS in most cases.

Abstract

Outside of shoebox rooms, acoustic diffraction phenomena are present and can influence important aspects of auditory perception, such as localisation. A simple extension of a shoebox room is an L-shaped room as it introduces a single diffracting edge. This paper presents two experiments carried out in L-shaped rooms in virtual reality. The first investigated whether the inclusion of diffraction modelling influences the perceived plausibility of the acoustic simulation, and the second to what extent newly developed efficient IIR filter diffraction models are equally plausible to the physically accurate Biot-Tolstoy-Medwin-Svensson (BTMS) model. The study compared diffraction of only the direct sound and diffraction of both direct and reflected sound. The results show that the inclusion of diffraction increased the perceived plausibility of the acoustic simulation. A statistically significant increase in plausibility was found by the addition of diffracted reflection paths, but only in the so-called shadow zone. The second experiment determined that the IIR filter diffraction models were similarly plausible to BTMS in 14 of 18 cases with a threshold of 0.5 on a 6-point Likert scale.

Figures