Modeling Speech using a Partially Observable Markov Decision Process

For over a decade, the Hidden Markov Model (HMM) has been the primary tool used for acoustic modeling in the field of speech recognition. In this dissertation we examine a more general approach using a Partially Observable Markov Decision Process (POMDP) to model the base phonetic unit. We introduce the concept of multiple phonetic context classes, one for each of the infinite possible contexts a phoneme can be in, and show how a POMDP can be used to represent such a model. Much the same way that tying mixtures at the state level across phonemes sharing linguistic properties is used to fill in gaps in the model space due to lack of data, the POMDP model can fill in additional gaps, in effect adding a second level of clustering driven by the data itself. An important result of this work is the ability of the POMDP model to represent the base phonetic unit, in all its possible contexts, within a single model. This enables us to look at the problem in a more abstract manner, allowing us to model better a phoneme to its fullest capability by developing relationships among its contexts that have thus far not been explored in current speech recognition research