A Parameterized Model of Psychoacoustical Roughness for Objective Vehicle Noise Quality Evaluation

ABSTRACT

In the assessment of vehicle noise quality, sound characteristics caused by modulation play an important role because they contribute significantly to the perceived annoyance. The proposed generalized model for the sensation of psychoacoustical roughness was developed as part of a research program with the aim of establishing an onboard analysis system for vehicle interior noise quality based on objective sound parameters.

The roughness model can be adjusted by model parameters to calculate versions of roughness, thereby accentuating different psychoacoustical assumptions. The model is based on the excitation-time pattern of the interior car sound which is recorded using an artificial head system. From the excitation-time pattern, a set of specific roughness parameters in overlapping critcal bands is obtained utilizing different weighting functions for the envelope spectrum in each band and various correlation methods between adjacent bands. The specific roughness parameters in critical bands are superimposed in several ways resulting in a set of modulation parameters. The model was successfully tested not only in predicting roughness assessment as reported in a companion paper, but it also proved to result in a valid objective parameter for the description of vehicle interior noise quality.

pdf File (92 KB)