Open Source WFST Tools for LVCSR Cascade Development
Josef R. Novak, Nobuaki Minematsu, Keikichi Hirose · 2011
This paper introduces the Transducersaurus toolkit which provides a set of classes for generating each of the fundamental components of a typical WFST-based ASR cascade, including HMM, Context-dependency, Lexicon, and Grammar transducers, as well as an optional silence class WFST. The toolkit further implements a small scripting language in order to facilitate the construction of cascades via a variety of popular combination and optimization methods and provides integrated support for the T 3 and Juicer WFST decoders, and both Sphinx and HTK format acoustic models. New results for two standard WSJ tasks are also provided, and the toolkit is used to compare a variety of construction and optimization algorithms. These results illustrate the flexibility of the toolkit as well as the tradeoffs of various build algorithms. 1