On the Steady-State Transmission in Rooms

Abstract

In 1935, E. C. Wente [J. Acoust. Soc. Am. 7, 123 (1935)] suggested an approach to the problem of finding a quantitative measure other than reverberation time for the acoustical quality of a room. The effect of reverberation time on acoustical quality was fairly well known at that time, and it had become apparent that reverberation time alone, though the most important room-acoustical parameter, was not sufficient to describe the acoustical behavior of an enclosure. Wente's approach consisted of measuring the steady-state transmission level from one point within a room to another as a function of frequency. He and other workers in the field suggested a variety of statistical quantities obtainable from such transmission curves as quality measures to supplement reverberation time. In this paper it will be shown that for large rooms these new parameters do not contain new information. In fact, the transmission curves can be proved to be members of an ensemble of statistical functions with a given amplitude distribution and a spectrum which depends only on reverberation time. Large rooms in the above sense are characterized by the fact that for each frequency in the range considered many normal modes are excited simultaneously. Assuming random distributions of the eigenfrequencies and the excitation coefficients [M. R. Schroeder, Acustica 4, Beih. 1, 456 (1954)] of the normal modes an inequality for the volume of the room can be derived which defines the range of validity of the statistical theory [M. R. Schroeder, Acustica 4, Beih. 2, 594 (1954)]. This theory has been confirmed by measurements in many concert halls, theaters, and broadcast studios [H. Kuttruff and R. Thiele, Acustica 4, Beih. 2, 614 (1954)].