A comparative study of methods for phonetic decision-tree state clustering

Phonetic decision trees have been widely used for obtaining robust context-dependent models in HMM-based systems. There are five key issues to consider when constructing phonetic decision trees: the alignment of data with the chosen phone classes; the quality of the modeling of the underlying data; the choice of partitioning method at each node; the goodness-of-split criterion and the method for determining appropriate tree sizes. A popular existing method usesefficient but crude approximate methods for each of these. This paper introduces and evaluates more detailed alternatives to the standard approximations.