论文信息 - Large Vocabulary Continuous Speech Recognition for Bulgarian

Large Vocabulary Continuous Speech Recognition for Bulgarian

The paper presents the results of a project completed by the authors for realizing a continuous speech recognition system for Bulgarian. The state-of-the-art speech recognition technology used in the system is discussed. Special attention is given to the problems with some speci cs of the Bulgarian language namely the large vocabulary (450000 wordforms). Some implementation details of the language module are given. At the end the paper provides evaluation of the accuracy of recognition.

[1] Bruce W. Watson,et al. Incremental construction of minimal acyclic finite state automata , 2000, CL.

[2] Lawrence R. Rabiner,et al. A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[3] Alex Acero,et al. Spoken Language Processing , 2001 .

[4] Hermann Ney,et al. On structuring probabilistic dependences in stochastic language modelling , 1994, Comput. Speech Lang..

[5] Mosur Ravishankar,et al. Efficient Algorithms for Speech Recognition. , 1996 .

[6] Fritz Hamm,et al. One-letter automata: How to reduce k tapes to one , 2008 .

[7] Gertjan van Noord,et al. Finite automata for compact representation of tuple dictionaries , 2004, Theor. Comput. Sci..

[8] Martin Kay,et al. Regular Models of Phonological Rule Systems , 1994, CL.

[9] Stephen E. Levinson,et al. Mathematical Models for Speech Technology , 2005 .