An Effective Low Complexity Binaural Beamforming Algorithm For Hearing Aids

In this paper, we propose a low complexity beamforming algorithm for binaural hearing aids that achieves a high degree of noise suppression without causing significant speech distortion. Specifically, we incorporate a priori approximate spatial information in the formulation and develop several effective techniques to reduce the communication overhead. For the resulting quadratically constrained beamforming problem, we propose a low complexity algorithm based on the Alternating Direction Method of Multipliers (ADMM) method. This new approach is shown to be computationally much more efficient than the previously proposed binaural beamforming algorithms. Through experiments on the real-world recording, the effectiveness of the binaural hearing aid design and the corresponding low complexity implementation are validated.