Speech enhancement using a perceptually constrained TQSVD-based approach incorporating PSO techniques

A new two-channel perceptually motivated subspace-based approach is proposed for the enhancement of speech corrupted by noise. The proposed approach takes into account the frequency masking properties of the human auditory system and reduces perceptual effects of the residual noise. The perceptually constrained Quotient Singular Value Decomposition (PCQSVD) method, uses the Particle Swarm Optimization (PSO) algorithm to estimate the additive noise. Very carefully performed objective evaluations and subjective tests show that the PCQSVD-PSO approach proposed here can offer improved speech quality, and intelligibility in the case of stationary and nonstationary noises, especially when the additive noise is nonwhite.