Speech enhancement using perceptual-decision-directed approach

摘要

Employing masking properties of the human ear to adapt a speech enhancement system has been widely developed to improve the quality of an enhanced speech signal. The spectral estimate of speech plays a major role in computing the noise masking threshold when simultaneous masking properties is considered. Although traditional methods using power-spectral-subtraction method to roughly estimate the speech spectra can provide acceptable performance, however, the estimation of speech spectra can be further improved for evaluating the noise masking threshold. In this article, we aim at finding a better spectral estimate of speech by two-step-decision-directed method. In turn, this estimate is employed to compute the noise masking threshold of a perceptual gain factor. Experimental results show that the background noise can be efficiently suppressed by embedding the two-step-decision-directed algorithm in the perceptual gain factor.