Why we should design for optimum reverberation times in rooms for speech communication

Abstract

Various texts list recommendations for optimum reverberation times as ideal goals in rooms for speech communication. However, some newer documents, such as ANSI S12.60, talk about maximum recommended reverberation times in rooms for speech communication. The source of the changed approach can be traced to interpretations of experimental results that do not consider the full implications for speech communication in real rooms, and also to the fact that reverberation time is not an ideal predictor of the quality of speech communication. In the extreme, minimizing reverberation times would lead to near anechoic rooms for speech and inadequate signal-to-noise ratios. The need for optimum reverberation times can be explained as a simple need to first achieve adequate signal-to-noise ratios. However, a more complete understanding is obtained by examining the benefits of early-arriving reflections of speech sounds on the intelligibility of the speech to listeners. Attempts to determine optimum reverberation times for normal hearing listeners, which are based on a balance between avoiding excessive reverberation and maintaining adequate signal-to-noise ratios, lead to a range of acceptable values that can vary with the ambient noise level. This discussion will be supported with examples from room acoustics measurements in classrooms and meeting rooms.