Joint Dereverberation and Beamforming With Blind Estimation of the Shape Parameter of the Desired Source Prior

Dereverberation and acoustic beamforming is used to capture the speech of a desired speaker in the presence of interfering speakers in a reverberant room using an array of microphones. Traditionally, to perform these two tasks, the desired speech is modelled in the time-frequency domain using a complex Gaussian (CG) prior with time-varying variances. The shape parameter of the prior distribution is fixed at the same value for all time-frequency bins. In this work, we propose to model the inverse of the variance (i.e. the precision parameter) of the CG prior distribution which controls the shape of the distribution as a Gamma distributed random variable. The hyperparameters of the Gamma distribution are then estimated based on the data captured by the microphones. This data-dependent blind estimation of the shape of the prior distribution helps the proposed algorithm to accurately model the desired speech and adapt to different speakers and acoustic scenarios better than algorithms with a fixed shape parameter. We use maximum likelihood techniques to estimate the multi-channel linear prediction (MCLP) dereverberation coefficients and the beamforming weights using the proposed signal model. The stochastically latent precision parameters are obtained by estimating the hyperparameters using the expectation maximization (EM) method. For the online version of the algorithm, a recursive EM method is also proposed for real-time processing. Extensive simulation results show improved dereverberation and interference cancellation performance of the proposed method highlighting the importance of not choosing the shape parameter of the prior distribution manually.