Enhanced Individualization of Head-Related Impulse Response Model in Horizontal Plane Based on Multiple Regression Analysis

One key issue in modeling head-related impulse responses (HRIRs) is how to individualize HRIRs model so that it is suitable for a listener. The objective of this research is to establish multiple regression models between minimum phase HRIRs and the anthropometric parameters in order to individualize a given listener's HRIRs with his or her own anthropometric parameters. We modeled the entire minimum phase HRIRs in horizontal plane of 37 subjects using principal components analysis (PCA). The individual minimum phase HRIRs can be estimated adequately by a linear combination of ten orthonormal basis functions. We proposed an enhanced individualization method based on multiple regression analysis of weights of basis functions by utilizing eight anthropometric parameters. Our objective simulation's results show that the estimated minimum phase HRIRs have small error and can be perceived similarly as the measured ones. In addition, the subjective localization performance of the estimated HRIRs is improved compared to the measured HRIRs.