A Speech Enhancement Algorithm Based on Robust Adaptive Spatial Beamforming

In order to further improve the noise reduction performance,the paper presents a robust adaptive spatial beamformer with post-filtering.The traditional beamformer tap delay-lines in the beamformer structure are replaced by microphones,thus making all microphones have a filter.The spatial beamformer gives rise to speech reference signal by classic linearly constrained minimum variance(LCMV),and the blocking matrix gives rise to noise reference signal.The speech reference and noise reference are across the adaptive multi-channel canceller to suppress disturbance noises effectively.The post-filtering technology is introduced to improve speech quality ulteriorly at last.The results show that compared with the traditional post-filtering adaptive beamforming algorithm,the proposed algorithm has better reduction performance and higher output signal to noise ratio.