14th International Congress of Phonetic Sciences (ICPhS-14)

San Francisco, CA, USA
August 1-7, 1999

An Improved Method of Automatically Generating Alternative Lexicon Pronunciations

Darryl Stewart, Philip Hanna, Ji Ming, F. Jack Smith

The Queens University of Belfast, UK

In recent years, various means of improving the quality of speech recognition lexicons by augmenting them with pronunciation variants and phonological rules have been explored. Alternative pronunciations can be generated using a Dynamic Programming (DP) approach, whereby different pronunciations of a word are compiled and any significant pronunciations extracted. However, such approaches do not compensate for co-articulation or Hidden Markov Model (HMM) misclassifications occurring in the observed phone sequence. Hence, the lexicon may contain poorly defined or spurious pronunciations.
   This paper proposes a novel means of automatically compiling pronunciation alternatives, through an approach which removes co-articulation effects and HMM misclassifications. This is accomplished through a probabilistic expansion of the observed pronunciations, using statistical information of how likely the observed pronunciation has been subjected to the aforementioned effects.
   The approach, when applied to the TIMIT database was found to improve the quality of the pronunciations generated using a DP based approach.

Full Paper

Bibliographic reference.  Stewart, Darryl / Hanna, Philip / Ming, Ji / Smith, F. Jack (1999): "An improved method of automatically generating alternative lexicon pronunciations", In ICPhS-14, 1713-1716.