Bisping R., Thoma G., Reichle M., Zeitler A.

The evaluation of car interior sounds in the laboratory may substantially differ from real life judgments of sound quality in the car. The main reason for this effect is the dynamic behavioural interaction of the driver with the process of sound generation, e.g. by accelerating the car, which does not occur in the laboratory. The driver, thus, gets an auditory feedback of his or her activity what may give a much more realistic impression of sound quality as compared to laboratory data. On the other hand, different from the laboratory, it is not easy for the experimenter to control for all sorts of disturbing environmental effects in the car. To overcome this problem the authors developed a standardized and reliable method for the modelling and validation of interior car sounds. Basically the method consists of a highly versatile synthesis device for online sound generation being controlled by various can-bus signals (load, rpm, speed, etc.) and a standardized experimental procedure for sound quality assessment. The paper will present the method and will give some experimental evidence for the validity of the approach. Auditory examples will be presented, too, demonstrating the high quality of the sound modelling procedure.