A Hybrid Algorithm for Predicting Median-Plane Head-Related Transfer Functions from Anthropometric Measurements

Xuejie Liu,Xiaoli Zhong,Hao Song

doi:10.3390/app9112323

Abstract

Since head-related transfer functions (HRTFs) represent the interactions between sounds and physiological structures of listeners, anthropometric parameters represent a straightforward way to customize (or predict) individualized HRTFs. This paper proposes a hybrid algorithm for predicting median-plane individualized HRTFs using anthropometric parameters. The proposed hybrid algorithm consists of three parts: decomposition of HRTFs; selection of key anthropometric parameters; and establishing a prediction formula. Firstly, an independent component analysis (ICA) is applied to median-plane HRTFs from multiple subjects to obtain independent components and subject-dependent weight coefficients. Then, a factor analysis is used to select key anthropometric parameters relevant to HRTFs. Finally, a regression formula that connects ICA weight coefficients to key anthropometric parameters is established by a multiple linear regression. Further, the effectiveness of the proposed hybrid algorithm is verified by an objective evaluation via spectral distortion and a subjective localization experiment. The results show that, when compared with generic Knowles Electronics Manikin for Acoustic Research (KEMAR) HRTFs, the spectral characteristics of the predicted HRTFs are closer to those of the individualized HRTFs. Moreover, the predicted HRTFs can alleviate front–back and up–down confusion and improve the accuracy of localization for most subjects.

Highlights

Three-dimensional spatial sound localization includes directional and distance localization
The combination of the objective standard deviation and the subjective localization experiment are used to verify the effectiveness of the hybrid algorithm based on anthropometric parameters for predicting individualized median-plane Head-related transfer functions (HRTFs)
In the proposed hybrid algorithm, independent component analysis is used to decompose the magnitude of HRTFs, factor analysis is used to select significant anthropometric parameters, and multiple linear regression is used to establish the prediction formula

Summary

Introduction

Three-dimensional spatial sound localization includes directional (i.e., azimuth and elevation) and distance localization. Lateral localization mainly depends on interaural differences, such as the interaural time difference (ITD) and the interaural sound level difference (ILD), caused by different transmission paths from sound sources to the left/right ear. On the other hand, mainly relies on the change in the magnitude spectrum when sound waves propagate to the ears, that is, the spectral cue. Head-related transfer functions (HRTFs), defined as the system transfer function of the sound filter from a sound source to human eardrums in the frequency domain, include the aforementioned ITD, ILD, and spectral characteristics. One major application of binaural HRTFs is the creation of a virtual auditory display (VAD). A VAD can be used as an experimental platform in research of binaural hearing, and has various applications in such fields as virtual reality, mobile communication, and multimedia [1,2,3].

Methods

Results

Discussion

Conclusion