Head-related transfer function (HRTF) synthesis based on a three-dimensional array model and singular value decomposition

An external ear model based on a three-dimensional array beamformer is presented to synthesize the head-related transfer functions (HRTFs). The array coefficients are calculated by matching the measured HRTFs with a frequency-domain template. In light of the characteristics of human hearing, a non-uniform sampling technique is also devised to reduce the problem to a manageable size. The model-matching problem is then solved, by using singular value decomposition procedure. Numerical simulations are carried out to investigate various array configurations in relation to approximation performance. A subjective experiment of sound localization is undertaken to demonstrate the validity of the proposed technique. From the results, the proposed beamformer-based technique proves to be an effective method for modeling the HRTFs.