Non-linear noise compensation for robust speech recognition using Gauss-Newton method

In this paper, we present the Gauss-Newton method as a unified approach to optimizing non-linear noise compensation models, such as vector Taylor series (VTS), data-driven parallel model combination (DPMC), and unscented transform (UT). We demonstrate that the commonly used approaches that iteratively approximate the noise parameters in an EM framework are variants of the Gauss-Newton method. Through the formulation of the Gauss-Newton method for estimating noise means and variances, the noise estimation problems are reduced to determining the Jacobians of the noisy speech distributions. For the sampling-based compensations, we present two methods, sample Jacobian average (SJA) and cross-covariance (XCOV), to evaluate the Jacobians. Experiments on the Aurora 2 database verify the efficacy of the Gauss-Newton method to these noise compensation models.