Claus Lynge Christensen (Odeon) – photo by Martin Arvebro
Room acoustic simulations and auralisation
Introduction text:
The aim of the workshop is to give the participants information on recent research in the field of room acoustic simulations and auralisations. The workshop will incorporate four key-note presentations that highlight interesting topics and applications in the field. The intension is that these presentations will inspire and engage to fruitful discussions during the workshop.
Moderator: Peter Svensson
Assistant moderator (Ecophon staff): Erling Nilsson
Opening lecture speaker (to all): Michael Vorländer
(Speakers in Work shop)
– Tapio Lokki
– Claus Lynge Christensen
– Peter Svensson
– Michael Vorländer
Workshop “Room Acoustic Simulations and Auralisation” at the Ecophon International Acousticians’ Seminar, 2011
Abstracts
Auralization of Sound Insulation
Michael Vorländer, RWTH Aachen University, Germany
Auralization is an established technique in room acoustic, but it is an interesting and powerful tool in other sub-disciplines of acoustics as well. In its usual definition, auralization means that a room is considered as system of sound transmission path and made “audible” by using excitation signals from appropriate sound sources. The key to achieve the resulting sound file is signal processing involving convolution with a filter response. The input data for such kind of auralization are based on standard building acoustics quantities such as sound reduction index, vibration reduction index and reverberation times, besides the basic room data of volume and surface areas. The problem in listening to auralized sound insulation is, however, the low level and low frequency content of the resulting signal which sets limits to the sensitivity and electrical noise in the electroacoustic reproduction system and to the background noise in the reproduction room. In this contribution the basic features of signal processing are discussed, and strategies for creating filters for direct and flanking sound transmission are presented and demonstrated with audio and video examples.
How to present auralized sound to a listener
Peter Svensson, Department of electronics and telecommunications Norwegian University of Science and Technology (NTNU)
Auralization involves the computation of a sound signal which then needs to be presented to a listener or “rendered”, to borrow a term from the visual domain. This presentation will give a brief overview of the common formats that are used. The most common is the binaural format, which is based on computing the sound pressure signals at the ears of a listener. Such a format can be presented to a listener using headphones, or two loudspeakers and a crosstalk cancellation filter. Another category includes various multichannel loudspeaker formats. These range from the 5.1 format which is used for home cinema, as well as more advanced ones: amplitude panning for any number of loudspeakers, wavefield synthesis, and higher-order Ambisonics. The latter two formats aim at reproducing a sound field over an extended volume in space so that a listener can move freely within a limited region. Possibilities and limitations with all the formats will be outlined.
Understanding acoustics of new trends in room design by means of auralisation
Claus Lynge Christensen, Odeon A/S
The design of buildings for schools, offices etc. have undergone large changes in recent years. Open plan design has become mainstream offering large and light, visually appealing atriums which unites complete buildings into large coupled spaces including many different functions. Even if room acoustics parameters such as T30 and STI (Speech Transmission Index) can be calculated, they may not offer a clear basis for evaluation of the acoustic comfort; if noise from the canteen is disturbing work in other parts of the building volume or if one group can hear the conversation of another (speech privacy) etc.
This presentation gives a short overview of the calculation principles in Odeon and demonstrates how Odeon was used for auralisation in the first open plan high school built according to the intensions of revised Danish guidelines (2005) for high schools. Auralisation was used in order to evaluate acoustics in the building with different amount of absorption present. Sound transmission through walls was also evaluated. Auralisation examples are given.
Modeling 1D-diffusers – the missing link.
Bengt-Inge Dalenbäck, CATT
Many surfaces in a room can in geometrical acoustics be seen as 1D-diffusers. Examples are stage or side walls with vertical battens or ridges, ceilings with beams across or along the room, 1D QRD diffusers, and in general 1D corrugated surfaces. So far these surfaces have been modeled as 2D-diffusers where the reflection angle depends only on the angle to the surface normal, such as in the Lambert distribution, and appears to have essentially worked satisfactory. However, the larger the area of surfaces of this type that are used in a room, and when source or receiver is close to these surfaces, the more important it becomes to use a more suitable scattering model, especially so if the room geometry is non-mixing and with an uneven absorption distribution. A 1D-diffuser can be described as a surface construction that, in a frequency dependent way, mainly scatters sound incident perpendicular to the battens while for sound incident parallel to the battens the scattering is low, but not as low as for a flat surface due to the depth variation. This paper presents a way to model 1D-diffusers that is compatible with the data for 2D-diffusers where only the direction of the battens needs to be indicated. The experience from estimating scattering coefficients for 2D-diffusers can be directly used and the model allows for both randomizing a diffused ray, as in classical ray-tracing, and for efficient deterministic diffuse ray split-up as used in CATT TUCT.
