We describe techniques for enhancing the accuracy, efficiency and features of a low-resource, medium-vocabulary, grammar-based speech recognition system, which uses hidden Markov models. Among the issues and techniques we explore are reducing computation via silence detection, applying the Bayesian information criterion (BIC) to build smaller and better acoustic models, minimizing finite state grammars, using hybrid maximum likelihood and discriminative models, and automatically generating baseforms from single new-word utterances. We report WER figures where appropriate.
Cite as: Deligne, S., Eide, E., Gopinath, R., Kanevsky, D., Maison, B., Olsen, P., Printz, H., Sedivy, J. (2001) Low-resource hidden Markov model speech recognition. Proc. 7th European Conference on Speech Communication and Technology (Eurospeech 2001), 1833-1836, doi: 10.21437/Eurospeech.2001-433
@inproceedings{deligne01_eurospeech, author={Sabine Deligne and Ellen Eide and Ramesh Gopinath and Dimitri Kanevsky and Benoit Maison and Peder Olsen and Harry Printz and Jan Sedivy}, title={{Low-resource hidden Markov model speech recognition}}, year=2001, booktitle={Proc. 7th European Conference on Speech Communication and Technology (Eurospeech 2001)}, pages={1833--1836}, doi={10.21437/Eurospeech.2001-433} }