Incorporating auditory properties into generalised sidelobe canceller

A novel speech enhancement method based on generalized sidelobe canceller (GSC) and auditory properties is presented. We show that it is possible to reduce audible speech distortions and preserve residual noise level under system model uncertainties. It can be done by constraining a speech leakage power according to masking effect phenomena. An optimal noise cancellation filter is derived using constrained minimization of the residual noise power. We implemented the GSC structure using a simple delay-and-sum beamformer and corresponding (delays-dependent) blocking matrix. Finally a comparative evaluation of the selected methods is performed using objective speech quality measures. The results show that the novel method outperforms conventional one providing lower speech distortions and comparable noise attenuation.