Use of Psychoacoustical Parameters for Soundscape measurement
Applied Soundscapes Symposium location:Manchester date:17 September 2009
The development of acoustical guidelines related to urban planning requires a simple measurement and evaluation method that could be accomplished by any acoustical consultancy office. However, a generally accepted, set of few acoustical descriptors for urban soundscapes that would be applicable for a guidelines has not been established yet.
This paper presents a simple method that is suitable for the classification of soundscapes of urban public places. The simplicity of the proposed method does not lie in the reduction of the description to one or a few numbers. In contrary, the followed approach allows to take into account as many parameters as judged relevant and necessary for an appropriate description. The parameters are used in an intermediate way in a classification algorithm that is based on the statistics of differences between a diversity of clusters of soundscapes in a substantially large set of recorded soundscape data.
The proposed classification algorithm processes a selection of characteristic acoustical features in a collection of sound fragments, recorded in streets, squares and parks, during so-called “soundwalks (SW)” that lasted for 10-15 minutes. The binaural recordings were done by means of in-ear microphones. The parameters that were used for clustering are based on statistical noise values (A-weighted) and psychoacoustical parameters (Loudness, Sharpness, Roughness and Fluctuation Strength). The statistical behavior of each parameter (Lx, Nx, Rx, Sx or Fx) is expressed by the values of the parameter that has been exceeded during a fraction of x % of the recording time, with the fraction typically taking the values 5% (exceptional events), 50% (probable situation) and 95% (quasi continuous situation). Also the so-called “urban interaural level difference“, a previously developed parameter, is included in the clustering analysis.
Given this objective classification of soundscapes into clusters or categories, the next research step will be to seek for correlations between the cluster structure on one hand, and the subjective categorization by people experiencing the respective urban public places on the other hand. If such a correlation could be established, this would open the way to design or adapt urban public places to match people’s expectations solely on the basis of the objective numbers.