Prediction of speech privacy between rooms

Abstract

The main purpose of sound insulation is usually to provide an appropriate level of speech privacy between rooms. However, the desired privacy is not always reached if, e.g. the background noise level of receiving room has not been considered or rooms are very small. This is often the case in dwellings and office rooms. Thus, sound reduction index does not correlate well with speech privacy. The aim of this study is to present a simple and validated model to predict the speech privacy between two arbitrary rooms when basic room and wall parameters are known. Speech privacy was described using Speech Transmission Index, STI. STI can be determined when the speech effort, room dimensions, early decay times of the rooms and background noise level of the reception room are known. The model was validated in several cases. The model is useful when economic optimization is appreciated. It can lead to major reductions or increments of conventional recommendations for Rw value, especially, when room dimensions are unusual or the rooms have special purposes. The optimization is very easy using a web-based software.