In this paper, we evaluate the performance of existing and new objective measures in terms of predicting the quality of reverberant speech and speech enhanced by dereverberation algorithms. We use subjective quality ratings designed to evaluate the quality of speech along three dimensions: speech coloration, reverberation tail effect and overall speech quality. Experimental results assess the correlations between the proposed objective quality measures and the three subjective rating scales and suggest that PESQ-based measures can very reliably predict the quality of reverberant and dereverberated speech.

In this work, we develop a two-stage blind dereverberation scheme for reverberant speech enhancement. The proposed algorithm operates by first splitting the reverberant inputs into different subbands. In the first stage, the inverse filters are estimated using the blind multiple input-output inverse-filtering theorem (MINT), while in the second stage our approach suppresses late reverberant energy by subtracting the power spectrum of the late impulse components from the power spectrum of the inverse filtered speech signal. Experimental results with reverberant speech indicate that the performance of the proposed algorithm is substantially better when compared to existing dereverberation algorithms.

In this letter we propose a novel approach for two-microphone enhancement of speech corrupted by reverberation. Our approach steers computational resources to filter coefficients having the largest impact on the error surface and therefore only updates a subset of coefficients in every iteration. Experimental results carried out in a realistically reverberant setup indicate that the performance of the proposed algorithm is comparable to the performance of its full-update counterpart.

This paper describes a highly practical blind signal separation (BSS) scheme operating on subband domain data to blindly segregate convolutive mixtures of speech. The proposed method relies on spatiotemporal separation carried out in the time domain by using a multichannel blind deconvolution (MBD) algorithm that enforces separation by entropy maximization through the popular natural gradient algorithm (NGA). Numerical experiments with binaural impulse responses affirm the validity and illustrate the practical appeal of the presented technique even for difficult speech separation setups.

In this contribution, we propose an entirely novel family of flexible score functions for blind source separation (BSS), based on the generalized Gamma family of densities. An efficient maximum likelihood (ML) technique for estimating the parameters of such score functions in an adaptive BSS setup, is also put forward. Simulations indicate that the proposed density model can approximate speech signals more accurately than conventional distributions, which leads to an increase in separation performance and convergence speed

This paper addresses the blind separation of convolutive and temporally correlated mixtures of speech, through the use of a multichannel blind deconvolution (MBD) method. In the proposed framework (LP-NGA), spatio-temporal separation is carried out by entropy maximization using the well-known natural gradient algorithm (NGA), while a temporal pre-whitening stage, based on linear prediction (LP), manages to fully preserve the original spectral characteristics of each source contribution. Confronted with synthetic convolutive mixtures, we show that the LP-NGA-an unconstrained natural extension to the multichannel BSS problem-benefits not only from fewer model constraints, but also from other factors, such as an overall increase in separation performance, spectral preservation efficiency and speed of convergence.

A number of commonly used methods for estimating the exponent parameter of a generalized Gaussian density (GGD) are reviewed, described and compared. More importantly, focusing on the family of entropy matching estimators (EMEs), a novel entropic expression with respect to higher-order moments of the modelled data is proposed. This yields an elegant generalized entropy matching estimator (G-EME). Comparative experimental results illustrate the high accuracy of the proposed estimator, for both light- and heavy-tailed distributions, as well as speech data.

This paper addresses the blind separation of convolutive and temporally correlated speech mixtures, through the use of a multichannel blind deconvolution (MBD) method. In the proposed method (NGA-LP) spatio-temporal separation is achieved by entropy maximization using the natural gradient algorithm (NGA), while a temporal prewhitening stage, based on linear prediction (LP), preserves the original spectral characteristics of each source contribution. It is further shown that a parameterized optimal nonlinearity derived from the generalized Gaussian density (GGD) model, increases the overall separation performance. Experiments with convolutive mixtures illustrate the merits of the proposed method.