Sixth International Conference on Spoken Language Processing
Modelling non-native speakers with different mother tongues is a difficult task for automatic speech recognition due to the large variation among speakers. One possibility for jointly modelling all speakers is to use the same speaker independent acoustic models and a joint lexicon to capture the variation. We have modified the reference lexicon using pronunciation rules that are derived in a totally data-driven manner from a set of adaptation data using the reference recognizer and the reference lexicon. Deriving common rules for such diverse sources simultaneously is difficult. The challenge is to combine these rules to a common set without increasing the confusability. In this paper we compare several methods of combining the individual rules to form a common lexicon for all speakers. Using a new log likelihood rule pruning measure presented in this paper, we achieved improved performance compared with more traditional rule pruning methods based on rule probability, and with much fewer rules. With a confusability reduction scheme we reduced the number of rules even further.
Bibliographic reference. Amdal, Ingunn / Korkmazskiy, Filipp / Surendran, Arun C. (2000): "Joint pronunciation modelling of non-native speakers using data-driven methods", In ICSLP-2000, vol.3, 622-625.