A theoretical bound for noise-robust speech recognition

Model compensation techniques for noise-robust speech recognition approximate the corrupted speech distribution. is work introduces a sampling method that, given speech and noise distributions and a mismatch function, in the limit calculates the corrupted speech likelihood exactly. For this, it transforms the integral in the likelihood expression, and then applies sequential importance resampling. ough it is too slow to compensate a speech recognition system, it enables a more ne-grained assessment of compensation techniques, based on the kl divergences to the ideal compensation for individual components. e kl divergence appears to predict the word error rate well. is technique alsomakes it possible to evaluate the impact of approximations that compensation schemes make. For example, this work examines the inžuence of the assumption that the corrupted speech distribution is Gaussian and diagonalising that Gaussian’s covariance. It also assesses the impact of a common approximation to the mismatch function for vts compensation, namely setting the phase factor to a xed value.