Abstract
The challenge in creating computational models for room acoustical auralizations lies in maintaining perceptual
credibility while keeping computational demand within bounds. Wave-based solvers can accurately describe
all wave phenomena but do so at an increasing computational cost for increasing frequencies. In this work, a
hybrid approach that combines the following parts is presented. Firstly, a time-domain wave-based method is
used for the low frequency range where the mesh can be coarse, at a limited computational demand. Secondly,
an image source model is combined with a diusion equation model for the high frequency range. Here the
image source model is used to calculate the early specular part of the impulse response, and the diusion
equation model is used to describe the late diuse part of the impulse response. This paper elaborates on
the details of this approach and its application to produce room impulse responses. A comparison between
room impulse responses created with this approach and those created with a reference wave-based solver for a
benchmark room is given.
credibility while keeping computational demand within bounds. Wave-based solvers can accurately describe
all wave phenomena but do so at an increasing computational cost for increasing frequencies. In this work, a
hybrid approach that combines the following parts is presented. Firstly, a time-domain wave-based method is
used for the low frequency range where the mesh can be coarse, at a limited computational demand. Secondly,
an image source model is combined with a diusion equation model for the high frequency range. Here the
image source model is used to calculate the early specular part of the impulse response, and the diusion
equation model is used to describe the late diuse part of the impulse response. This paper elaborates on
the details of this approach and its application to produce room impulse responses. A comparison between
room impulse responses created with this approach and those created with a reference wave-based solver for a
benchmark room is given.
Original language | English |
---|---|
Title of host publication | Proceedings of Euronoise 2021, Online, 2021 |
Publication status | Published - 25 Sept 2021 |