ITERATIVE AND SEQUENTIAL KALMAN FILTER-BASED SPEECH ENHANCEMENT ALGORITHMS

Speech quality and intelligibility might significantly deteriorate in the presence of background noise, especially when the speech signal is subject to subsequent processing. In particular, speech coders and au tomatic speech recognition (ASR) systems that were designed or trained to act on clean speech signals might be rendered useless in the prese nce of background noise. Speech enhancement algorithms have therefore attracted a great deal of interest in the past two decades. Zn this pa per, we present a class of Kalman filter-based algorithms with some ex tensions, modifications, and improvements of previous work. The first algorithm employs the estimate-maximize (ER I) method to iteratively e stimate the spectral parameters of the speech and noise parameters. Th e enhanced speech signal is obtained as a byproduct of the parameter e stimation algorithm. The second algorithm is a sequential, computation ally efficient, gradient descent algorithm. We discuss various topics concerning the practical implementation of these algorithms. Extensive experimental study using real speech and noise signals is provided to compare these algorithms with alternative speech enhancement algorith ms, and to compare the performance of the iterative and sequential alg orithms.