Research on the sound metric of door-slamming sound based on leaky integration and wavelet decomposition

The evaluation of the sound quality of door-slamming has become one of the important issues in vehicle noise, vibration and harshness (NVH) analysis. For the sound quality evaluation of door-slamming, a new sound metric, named as sound metric based on critical band wavelet decomposition (SMCBWD), is developed. In the new sound metric, the sound signals of door-slamming are sampled and the signal component of the door-slamming sound which has the great influence on the quality of door-slamming is extracted by using the leaky integration method. The extracted signal component is then decomposed by wavelets based on the critical bands and the coefficients of wavelet decomposition are calculated. Based on the energy of the frequency weighted wavelet decomposition coefficients, the new sound metric, SMCBWD, is calculated. In order to verify the effectiveness of SMCBWD, the correlation coefficients between the new sound metric and the subjective sound quality performance value of door-slamming, as well as between the traditional sound metrics (loudness, sharpness) and the subjective sound quality performance value of door-slamming have been calculated, respectively. The results show that the new sound metric developed in this paper has the higher correlations with the subjective sound quality performance value when compared with the traditional sound metric of loudness and sharpness. Thus, SMCBWD can be used to evaluate the sound quality of door-slamming more accurately.