A METHOD FOR MODELING HEAD-RELATED TRANSFER FUNCTION FOR IMPLEMENTING 3D VIRTUAL SOUND, AND METHOD AND APPARATUS FOR IMPLEMENTING 3D VIRTUAL SOUND USING THE SAME

摘要

A method for modeling a head-related transfer function for implementing 3-D virtual sound, and a method and an apparatus for implementing 3-D virtual sound using the same are provided to simultaneously implement decoding of an audio stream and 3-D virtual sound composite. Each time domain HRTF(Head Related Transfer Function) is converted into a filter bank of a subband domain by configuring equivalent subband domain HRTF filter matrixes. Elements of the HRTF filter matrixes are grouped into plural sets. Plural principal vectors are extracted from each set, and then weights of the principal vectors corresponding to the elements of each set are obtained. In the step of obtaining the filter bank of the subband domain, each time domain HRTF corresponding to different positions of 3-D space is converted into the subband domain HRTF filter matrix.