Quantitative Perceptual Separation of TwoKinds of Degradation in Speech Denoising Applications

摘要

Classical objective criteria evaluate speech quality using one quantity which embed all possible kinds of degradation. For speech de-noising applications, there is a great need to determine with accuracy the kind of the degradation (residual background noise, speech distortion or both). In this work, we propose two perceptual bounds UBPE and LBPE denning regions where original and denoised signals are perceptually equivalent or different. Next, two quantitative criteria PSANR and PSADR are developed to quantify separately the two kinds of degradation. Some simulation results for speech denoising using different approaches show the usefulness of proposed criteria.