Abstract

The desired beam pattern of the array can be achieved by applying specific array weighting in the eigenbeam space derived from the plane wave decomposition. Head related transfer function (HRTF) data are selected automatically from the KEMAR HRTF measurements database of MIT media lab without the message of the direction of the incoming sound wave. The HRTF for the sound wave direction are derived by combining the directivity of the eigenbeam beamforming and the algebraic expression of HRTF. Computer simulations demonstrate that the HRTF approximation derived from this paper are similar to those of the measurements by MIT media lab in certain frequency band. The higher the order of the eigenbeam used in eigenbeam beamforming, the more similar of the derived HRTF data with those of the measurements. Listening experiments are conducted to evaluate the efficiency of the proposed HRTF auto-selection algorithm subjectively. The results indicate that the audio localization precision for virtual sounds using approximated HRTF is consistent with those of the measured HRTF, which verifies the validity of the proposed algorithm.

This work is licensed under a Creative Commons Attribution 4.0 License.