An N-best strategy, dynamic grammars and selectively trained neural networks for real-time recognition of continuously spelled names over the telephone
作者： J.-C. Junqua , S. Valente , D. Fohr , J.-F. Mari
DOI: 10.1109/ICASSP.1995.479828
关键词: Feature (machine learning) 、 Computer science 、 Speech processing 、 Speech recognition 、 Artificial neural network 、 Rule-based machine translation 、 Artificial intelligence 、 Hidden Markov model 、 Grammar 、 Natural language processing
摘要: We introduce SmarTspelL, a new speaker-independent algorithm to recognize continuously spelled names over the telephone. Our method is based on an N-best multi-pass recognition strategy applying costly constraints when number of possible candidates low. This outperforms HMM recognizer using grammar containing all names. It also more suitable real-time implementation. For 3388 name dictionary, 95.3% rate obtained. A prototype has been implemented workstation. present comparisons different feature sets for speech representation, and two approaches first- second-order HMMs.
sci-hub.se 参考文章(8)
Hynek Hermansky, Phil Kohn, Nelson Morgan, Aruna Bayya, Compensation for the effect of the communication channel in auditory-like analysis of speech (RASTA-PLP). conference of the international speech communication association. ,(1991)
Jean-Claude Junqua, Dominique Fohr, Yolande Anglade, Jean-François Mari, Hidden Markov models and selectively trained neural networks for connected confusable word recognition. conference of the international speech communication association. ,(1994)
Y. Anglade, D. Fohr, J.-C. Junqua, Speech discrimination in adverse conditions using acoustic knowledge and selectively trained neural networks IEEE International Conference on Acoustics Speech and Signal Processing. ,vol. 2, pp. 279- 282 ,(1993) , 10.1109/ICASSP.1993.319290
Hy Murveit, John Butzberger, Mitch Weintraub, Reduced channel dependence for speech recognition Proceedings of the workshop on Speech and Natural Language - HLT '91. pp. 280- 284 ,(1992) , 10.3115/1075527.1075593
Richard Schwartz, Steve Austin, Efficient, high-performance algorithms for N-Best search human language technology. pp. 6- 11 ,(1990) , 10.3115/116580.116581
J.-F. Mari, J.-P. Haton, A. Kriouile, Automatic word recognition based on second-order hidden Markov models IEEE Transactions on Speech and Audio Processing. ,vol. 5, pp. 22- 25 ,(1997) , 10.1109/89.554265
Krist Roginski, Mark A. Fanty, Ronald A. Cole, English alphabet recognition with telephone speech. conference of the international speech communication association. ,(1991)
Climent Nadeu, Biing-Hwang Juang, Filtering of spectral parameters for speech recognition conference of the international speech communication association. ,(1994)