Pronunciation Modeling and Lexicon Adaptation for Spoken Language Technology (PMLA) September 14-15, 2002 Aspen Lodge, Estes Park, Colorado, USA

Pronunciation Modeling and Lexical Adaptation Using Small Training Sets

Louis ten Bosch (1), Nick Cremelie (2)

(1) University of Nijmegen, The Netherlands
(2) ScanSoft BVBA., Merelbeke, Belgium

A method for data-driven lexical adaptation on the basis of a limited number of acoustic training tokens is discussed. The method is closely related to pronunciation modeling techniques. A set of pronunciation variants is generated by forced alignment, followed by a step to select promising pronunciation candidates by using a ranking function. The method has been validated on a database consisting of short utterances (proper names) spoken by native and non-native speakers. In the case of 5 training tokens per word, an improvement of 10-30 percent relative could be obtained compared to the baseline. A number of possible improvements of this method are discussed as well.

Full Paper

Bibliographic reference.  Bosch, Louis ten / Cremelie, Nick (2002): "Pronunciation modeling and lexical adaptation using small training sets", In PMLA-2002, 111-116.