Title :
Stochastic lexicon modeling for speech recognition
Author :
Yun, Seong-Jin ; Oh, Yung-Hwan
Author_Institution :
Dept. of Comput. Sci., Korea Adv. Inst. of Sci. & Technol., Taejon, South Korea
Abstract :
To optimally cope with continuous speech recognizer, we propose the stochastic lexicon model that effectively represents variations in pronunciation. In this lexicon model, the baseform of a word is represented by subword-states with a probability distribution of subword units as a two-level hidden Markov model (HMM) and this baseform is automatically trained by sample utterances. Also, the proposed approach can be applied to systems employing nonlinguistic recognition units.
Keywords :
hidden Markov models; probability; speech recognition; stochastic processes; HMM; continuous speech recognition; nonlinguistic recognition units; probability distribution; pronunciation variations; sample utterances; speech recognition; stochastic lexicon modeling; subword unit; subword-states; two-level hidden Markov model; Automatic speech recognition; Computer science; Dictionaries; Error analysis; Hidden Markov models; Probability distribution; Speech processing; Speech recognition; Stochastic processes; Vocabulary;
Journal_Title :
Signal Processing Letters, IEEE
