AES London 2011
Poster Session P19
P19 - Room Acoustics
Monday, May 16, 09:00 — 10:30 (Room: Foyer)
P19-1 System Identification of Equalized Room Impulse Responses by an Acoustic Echo Canceller Using Proportionate LMS Algorithms—Stefan Goetze, Feifei Xiong, Fraunhofer Institute for Digital Media Technology IDMT - Ilmenau, Germany; Jan Ole Jungmann, University of Lübeck - Lübeck, Germany; Markus Kallinger, University of Oldenburg - Oldenburg, Germany (now with Franhofer Institute for Integrated Circuits IIS, Erlangen, Germany); Karl-Dirk Kammeyer, University of Bremen - Bremen, Germany; Alfred Mertins, University of Lübeck - Lübeck, Germany
Hands-free telecommunication systems usually employ subsystems for acoustic echo cancellation (AEC), listening-room compensation (LRC), and noise reduction in combination. This contribution discusses a combined system of a two-stage AEC filter and an LRC filter to remove reverberation introduced by the listening room. An inner AEC is used to achieve initial echo reduction and to perform system identification needed for the LRC filter. An additional outer AEC is used to further reduce the acoustic echoes. The performance of proportionate filter update schemes such as the so-called proportionate normalized least mean squares algorithm (PNLMS) or the improved PNLMS (IPNLMS) for system identification of equalized impulse response (IR) are shown and the mutual influences of the subsystems are analyzed. If the LRC filter succeeds in shaping a sparse overall IR for the concatenated system of LRC filter and room impulse response (RIR) the PNLMS performs best since it is optimized for the identification of sparse IRs. However, the equalization may be imperfect due to channel estimation errors in periods of convergence and due to the so-called tail-effect of AEC, i.e., the fact that only the first part of a RIR is identified in practical systems. The IPNLMS is more appropriate in this case to identify the equalized IR.
Convention Paper 8437 (Purchase now)
P19-2 Virtual Room Acoustics : A Comparison of Techniques for Computing 3-D-FDTD Schemes Using CUDA—Craig Webb, Stefan Bilbao, University of Edinburgh - Edinburgh, Scotland, UK
High fidelity virtual room acoustics can be approached through direct numerical simulation of wave propagation in a defined space. Three-dimensional Finite Difference Time Domain schemes can be employed and adept well to a parallel programming model. This paper examines the various approaches for calculating these schemes using the Nvidia CUDA architecture. We test the different possibilities for structuring computation, based on the available memory objects and thread-blocking model. A standard test simulation is computed at double precision under different arrangements. We find that a 2-D extended tile blocking system, combined with shared memory usage, produces the fastest computation for our scheme. However, shared memory usage is only marginally faster than direct global memory access, using the latest FERMI GPUs.
Convention Paper 8438 (Purchase now)
P19-3 Acoustic Parameters of Chosen Orthodox Churches Overview and Preliminary Psychoacoustic Tests Using Choral Music—Pawel Maleck, Jerzy Wiciak, AGH University of Science and Technology - Kraków, Poland
A lot of acoustic research was done for Roman Catholic churches and because of some differences in traditions and culture comparing to Orthodoxy, it is not the best idea to use its acoustic estimators for Orthodox churches. The paper shows results of measurements in some Orthodox churches in Poland and the proposal of psychoacoustic tests using the convolution technique, which would allow formulating the new acoustic outlines for Orthodox churches. The research has been done especially considering choral music, which is an inseparable part of Eastern Christian culture; so as a test, there were Orthodox choir sound samples recorded in an anechoic chamber convoluted with impulse response of measured churches.
Convention Paper 8439 (Purchase now)
P19-4 A Comparative Study of Various “Optimum” Room Dimension Ratios—John Sarris, Aretaieio University Hospital - Athens, Greece
Various “optimum” room dimension ratios that have been proposed in the literature are studied and compared. Since each proposal is based on a different criterion, independent objective measures of the acoustic quality of the various room ratios are applied in this paper. The most straightforward metric is the flatness of a room’s corner to corner frequency response, but since this is not representative of the variations of the sound pressure within the closed space, different metrics are employed to quantify these variations. Simulation results are presented to evaluate the effectiveness of the various ratios for the case of a small and a larger room.
Convention Paper 8440 (Purchase now)
P19-5 Perception of Spatial Distribution of Room Response Reproduced Using Different Loudspeaker Setups—Javier Gomez, Aalto University - Espoo, Finland; Rafael C. D. Paiva, Aalto University - Espoo, Finland, Nokia Technology Institute INdT, Brazil; Kai Saksela, Thomas Svedström, Ville Pulkki, Aalto University - Espoo, Finland
A listening test was conducted to assess the effect that different direct-to-reverberant ratios, loudspeaker setups, and reverberation times have on the directional perception of a synthetic room response. The results show that the perceived spatial distribution of reverberation is dependent on the speaker setup, on the reverberation time, and on the listener. Additionally, they show that a small amount of direct sound, even when barely masked by reverberation, shifts the overall perception to the front.
Convention Paper 8441 (Purchase now)
P19-6 Modal Analysis and Sound Field Simulation of Vibration Panels in a Free Sound Field and in a Rectangular Enclosure—Kazuhiko Kawahara, Akihiro Sonogi, Kyushu University - Fukuoka, Japan; Shin-ya Sato, Kyushu University - Fukuoka, Japan, Nittobo Acoustic Engineering Co., Ltd., Tokyo, Japan
Several types of implementation of panel loudspeakers, for example, distributed mode loudspeaker (DML), are proposed. The diaphragm of a DML is thought to behave like a bending plate. The diaphragm of an electrostatic loudspeaker is thought move as a rigid plate. What is the difference of an acoustical sound field generation feature with each implementation? In this paper we used models of same dimensional rectangular panel. One is a rigid diaphragm. Another is a bending panel that has an actuator in the center of it. We made simulations of directivity and sound pressure distribution in a rectangular room. We made simulations of sound pressure distribution in a rectangular room. We could show less coherent SPL distribution in the case of a bending panel.
Convention Paper 8442 (Purchase now)
P19-7 Simple Room Acoustic Analysis Using a 2.5 Dimensional Approach—Patrick Macey, PACSYS Limited - Nottingham, UK
Cavity modes of a finite bounded region with rigid boundaries can be used to compute the steady state harmonic response for excitation of an acoustic cavity by a point source. In Cuboid domains, with analytical modes, this is straightforward. For more general regions, determining a set of orthonormal modes is more difficult. The finite element method could be used for arbitrary 3-D regions. The current work investigates a hybrid numerical/analytical method, applicable to rooms of constant height, but arbitrary cross section. A 2-D FE analysis is used to compute the cross-section modes/frequencies. Three-dimensional modes are constructed as an outer product of the cross section modes and 1-D modes for the height. Comparison is made with BEM computations.
Convention Paper 8443 (Purchase now)
P19-8 Early Reflections Design for Natural Stereo Sound Listening—Chul-Jae (Jay) Yoo, RADSONE, Inc. - Gyeonggi-do, Korea
For natural stereo sound in rooms, some considerations must be followed. For example, to maximize time density, Feedback Delay Network (FDN) is used in which the feedback matrix has every element consisting of absolute value 1. To model faster high frequency decay according to the increasing number of reflections, absorption filters after delay lines in the FDN are generally used. Mixing matrix for uncorrelated L/R output is also used. In this paper an elaborate design scheme of the first early reflection portion in early reflections was proposed to obtain higher IACC values than those of conventional systems, resulting in a more precise sound source image with ambience.
Convention Paper 8444 (Purchase now)