Decision tree clustering for KL-HMM
David Imseng, John Dines · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · 2012
Recent Automatic Speech Recognition (ASR) studies have shown that Kullback-Leibler diverge based hidden Markov models (KL-HMMs) are very powerful when only small amounts of training data are available.However, since the KL-HMMs use a cost function that is based on the Kullback-Leibler divergence (instead of maximum likelihood), standard ASR algorithms such as the commonly used decision tree clustering are not applicable in general.In this communication, we present an algorithm that allows us to perform decision tree clustering for KL-HMM based ASR systems.