Chapter 2 Computational System of Simulating the Sound Field in a Room

2.1. Introduction

    This paper describes a procedure for the acoustic design of a concert hall that applies the theory of subjective preference, which is associated with the human cerebral hemispheres. Four orthogonal physical factors can be calculated using architectural drawings.

2.2. Outline of Simulating the Sound Field in a Room
    The testing room is divided into a sound field simulation room (Figure 2.1) and a control room (Figure 2.2), which are partitioned by a double soundproof door. The interior of the room is decorated and illuminated so that the user (the subject who is being tested) does not feel locked inside a special room, surprised at being in a soundless state that is not ordinarily experienced. In the control room, whose functions are described below, three workstations and a sound field synthesizer (Figure 2.3) are installed on a free-access floor. The test listening method is a multi-speaker method that allows up to four people to listen simultaneously. The speakers are arranged to emphasize the precise reproduction of the direction of reflected sound. The entire surface of the speakers is covered with glass wool to a thickness of 50 mm, minimizing the reflection of medium- and high-frequency sound.

Figure 2.1. The souond simulation room of the Miyama Conceru. Reproduced with kind permission of Mr Suiyo Sato, Suikoh-sya, Tokyo, Japan.

Figure 2.2. The control  room for sound simulation etc. Reproduced with kind permission of Mr Suiyo Sato, Suikoh-sya, Tokyo, Japan.

Figure 2.3. The hardware for the souond simulation. Reproduced with kind permission of Mr Suiyo Sato, Suikoh-sya, Tokyo, Japan.

Composition of the overall software system
    The system shown in Figure 2.4 is a computer system based on the client-server model and operate under three different operating systems, namely, DEC VMS, NEC OS-2, and Windows. The acoustical CAD system supports everything from the input of architectural drawings to the acoustical design of the interior of the room by the sound ray tracing method and mirror image method, as well as its visual presentation. The part of the routine of the effective wall is shown in Appendix A. A file server allows the sound field auralization system [47, 48] to use the results of the acoustical CAD system. The auralization system also feeds the results to the devices for sound field synthesis and equalization. The seat reservation and ticket issuing system, as an independent subsystem on the LAN, has the functions of event control, testing for sound preferences, issuance of individual cards, seat selection, ticket reservation, and ticket issuance. These functions can run simultaneously from multiple (at present, two) workstations on the LAN.

Figure 2.4. Composition of the overall software and hardware system.

2.3. An Application in Designing of the Kirishima International Concert Hall
    The physical factors at each seat were calculated at the design stage with architectural drawings using the image method [31]. In the computer mode, the effect of diffusers on the walls at high frequencies, reflections and scattering caused by seats were not taken into account in the calculation. A single source was considered, located at 3m from the stage front, 2m to the side of the centre line, and 1.5 m above the stage Boor. The direct sound and reflections up to third order, which included about 300 reflections, with their amplitudes, delay times, and directions of arrival at the listeners were calculated. In the calculation of IACC, white noise was selected as the source signal. In this calculation, the reverberation time is assumed to be 1.8 s (500Hz, occupied) throughout the hall.
    The calculated results for the physical factors of the listening level, the initial time delay gap, the IACC, and the total amplitude of reflections (A-Value) are shown in Figure 2.5. The value for the listening level has a range of 4.0 dB through b out the main audience moor. The largest initial time delay Dt1 (27 ms) is observed near the source on the center line of the hall. The IACC at each seat is less than 0.2, except for seats close to the source.
    Scale values of subjective preference for listeners are shown in Figure 2.6. For slow tempo music (Music motif A), positioning the source near to the stage front increases the total scale value in seats near the back wall of main audience moor. For fast tempo music (Music motif B), positioning the source near the back wall of the stage increases the total scale value in seats at the center of the main audience floor. As far as the type of music is concerned, the fast tempo music may be best suited for this hall. Preference for performers is not discussed here, but it has been obtained for changes in the delay time of a single reflection [32].

Figure 2.5. Results of calculated values of physical factors: (a) Contour lines of equal relative listening level [dB]; (b) Contour lines of initial time delay gap between the direct sound and first reflection excluding the floor reflection [ms] ; (c) Contour lines of equal IACC for white noise; (d) Contour lines of equal A-value.

Figure 2.6. Results of calculated scale values of subjective preference for two source positions: (a) Music motif A; (b) Music Motif B.

2.4. Remarks
    Based on previous study, sound fields ware simulated by the concert hall building plan in the designing stage. After obtaining the binaural impulse responses at each seat, four orthogonal factors including the sound pressure level (SPL), the initial time delay gap between the direct sound and the first reflection (Dt1), the subsequent reverberation time (Tsub) and the interaural crosscorrelation (IACC) ware analyzed to evaluate subjective preference. The sound fields at each seat in the concert hall can be evaluated before construction.