This paper describes the results of the objective measurements and the subjective assessments on the acoustical characteristics of the two large music studios of NHK. They have similar room shape, volume and reverberation but have different reputations. It also describes the desirable acoustical properties at the microphone positions in music studios suggested by these results. It is concluded that; (a) The size of the definition D and log ratio of early to reverberant sound energy R at the microphone positions are related to the subjective assessment of liveness. (b) The frequency characteristics of R are related to the assessments of the naturalness or liveness, ballance of the sounds of each instruments, etc. (c) The desirable R at single pick-up positions is about-2〜3dB, and its value at multiple pick-up positions is about 10〜16dB. (d) In acoustical planning of a large music studio, setting of the stage reflectors and the finishing treatment of "live end-dead end" type are the safest policy.
A new method based on spatial correlation techniques to evaluate the degree of diffuseness of the sound field in a reverberation chamber was described. Directional distribution of correlation coeffcients R_π were observed under various conditions in reverberation chambers, and a diffusitivity index R_π, was defined. Close dependence between R_π and measured reverberant absorption coefficients of the material was recognised. For the reverberation chamber in which R_π, is greater than 70%, it could be judged that its sound field would fulfil the requisite for the diffuse sound field, this being the premise of the reverberation chamber method.
When room acoustics are analyzed by acoustical scalemodel the model room must be set to a situation equivalent to the actual room and to wall material of equivalent acoustic impedance, for the law of similarity between actual and its model room to hold. In the first place the author made use of the material of equivalent sound absorption coefficient instead of equivalent acoustic impedance. The author, therefore, made a model-reverberation room, and tried to find the sound absorption coefficient of the wall material in a 1/10 scale-model room. In order to reduce the sound attenuation due to humidity in the air in high frequencies, the author studied the law of similarity between reverberation time and reduction scale of the model, using the relation between reverberation time and attenuation coefficient due to humidity in the air. As the result, it has been proved that dehydration in model room is effective. Reverberation time was deter-mined by using a high speed tape recorder that recorded reverberation sound in the model room with dry air, and reproducing it by reducing the tape speed to 1/10. By way of example, the author made 1/10 scale-models of a chapel and a hall, and reported the results on the reverberation time, sound pressure response and echo pattern as compared with the actual room.
This paper describes the criterion of the perceptibility of the echo which is necessary for simply detecting the echo due to multi-reflections by the observation of echo time diagrams The perceptible limit of the echo is given by a single curve where the ordinate is the listening sound level and the abscissa is the time delay of the reflection. Then, (a) the inclination of the perceptible time level characteristic of the echo is given, (sound pressure level)/ (300 milli-seconds), (b) in the case of the reflection, where the level is larger than that of the perceptible limit existing, the masking of the echo begins from that point again, (c) the perceptibility of the echo is concerned with the projecting level over the perceptible limit. The projecting level in which echoes are slightly audible is within 1. 5〜4dB high in the range of the frequency of 500-2000 Hz.
In this paper, simulation of the sound attenuation caused by air absorption is studied as to acoustic scale model experiment of room acoustics. Examining the data of sound absorption by air published up to this time and measured by the authors, it is found that in 1/10 scale model experiment the sound absorption by air under normal room condition can be simulated by excluding the molecular absorption of air and reducing the absorption by experimental medium gas to the value of the classical absorption of air. Thus, N2-substitution method proved useful for simulation.