Analysis of sound characteristics to design an annoyance metric for rattle sounds in the automotive industry

Abstract

In this research, a new objective psychoacoustic metric for predicting the annoyance level of the rattle sounds in the automotive industry is proposed. Compared to previous works, in addition to the standard psychoacoustic metrics, new statistical measures of average relative prominence and normalised number of peaks are introduced and employed. A subjective listening test was conducted to elicit the perception of annoyance level of the rattle sounds by the car users. The subjective test was based on the new method of paired comparison with indirect magnitude estimation, developed and proposed in this study by the authors. Accuracy of the metric proposed was checked by the coefficient of determination and error measures. The perceived annoyance metric proposed can be tuned for predicting other similar impulsive sounds in other disciplines as well. Also, the authors would like to encourage other researchers to employ the subjective test method proposed for other similar applications.