Speech enhancement using nonlinear microphone array based on complementarybeamforming

This paper describes a spatial spectral subtraction method by using the com plementary beamforming microphone array to enhance noisy speech signals for speech recognition. The complementary beamforming is based on two types of beamformers designed to obtain complementary directivity patterns with res pect to each other. In this paper, it is shown that the nonlinear subtracti on processing with complementary beamforming can result in a kind of the sp ectral subtraction without the need for speech pause detection. In addition , the optimization algorithm for the directivity pattern is also described. To evaluate the effectiveness speech enhancement experiments and speech re cognition experiments are performed based oil computer simulations under bo th stationary and nonstationary noise conditions. In comparison with the op timized conventional delay-and-sum (DS) array, it is shown that: (1) the pr oposed array improves the signal-to-noise ratio (SNR) of degraded speech by about 2 dB and performs more than 20% better in word recognition rates und er the conditions that the white Gaussian noise with the input SNR of -5 or -10 dB is used, (2) the proposed array performs more than 5% better in wor d recognition rates under the nonstationary noise conditions. Also, it is s hown that these improvements of the proposed array are same as or superior to those of the conventional spectral subtraction method cascaded with the DS array.