This paper describes how we can use the generalized Baum-Welch (GBW) algorithm to develop better extended Baum-Welch (EBW) algorithms. Based on GBW, we show that the backoff term in the EBW algorithm comes from KL-divergence which is used as a regularization function. This finding allows us to develop a fast EBW algorithm, which can reduce the time of model space discriminative training by half, without incurring any degradation on recognition accuracy. We compare the performance of the new EBW algorithm with the original one on various large scale systems including Farsi, Iraqi and modern standard Arabic ASR systems.
Bibliographic reference. Hsiao, Roger / Schultz, Tanja (2011): "Generalized Baum-welch algorithm and its implication to a new extended Baum-welch algorithm", In INTERSPEECH-2011, 773-776.