A single channel speech enhancement technique exploiting human auditory masking properties

[摘要] To enhance extreme corrupted speech signals, an Improved PsychoacousticallyMotivated Spectral Weighting Rule (IPMSWR) is proposed, that controls thepredefined residual noise level by a time-frequency dependent parameter.Unlike conventional Psychoacoustically Motivated Spectral Weighting Rules(PMSWR), the level of the residual noise is here varied throughout theenhanced speech based on the discrimination between the regions with speechpresence and speech absence by means of segmental SNR within critical bands.Controlling in such a way the level of the residual noise in the noise onlyregion avoids the unpleasant residual noise perceived at very low SNRs. Toderive the gain coefficients, the computation of the masking curve and theestimation of the corrupting noise power are required. Since the clean speechis generally not available for a single channel speech enhancement technique,the rough clean speech components needed to compute the masking curve arehere obtained using advanced spectral subtraction techniques. To estimate thecorrupting noise, a new technique is employed, that relies on the noise powerestimation using rapid adaptation and recursive smoothing principles. Theperformances of the proposed approach are objectively and subjectivelycompared to the conventional approaches to highlight the aforementionedimprovement.

[学科分类] 电子、光学、磁材料