System and method using N-best strategy for real time recognition of continuously spelled names
作者： Jean-Claude Junqua
DOI: 10.1121/1.423099
关键词:
摘要: A multipass recognition strategy selects the N-best hypotheses resulting from each pass and propagates these to next pass. This outperforms conventional hidden Markov model recognizers using a grammar constraining all possible names. Real time of continuously spelled names is made feasible, in part, because processor-intensive costly constraints are applied, if at all, 4th pass, after system has produced much smaller dynamic grammar.
lens.org 参考文章(18)
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)
J.-C. Junqua, S. Valente, D. Fohr, J.-F. Mari, An N-best strategy, dynamic grammars and selectively trained neural networks for real-time recognition of continuously spelled names over the telephone international conference on acoustics, speech, and signal processing. ,vol. 1, pp. 852- 855 ,(1995) , 10.1109/ICASSP.1995.479828
R. Schwartz, Y.-L. Chow, The N-best algorithms: an efficient and exact procedure for finding the N most likely sentence hypotheses international conference on acoustics, speech, and signal processing. pp. 81- 84 ,(1990) , 10.1109/ICASSP.1990.115542
Raymond Lau, Salim Roukos, Ronald Rosenfeld, Building scalable N-gram language models using maximum likelihood maximum entropy N-gram models ,(1995)
Takao Watanabe, Pattern matching apparatus ,(1983)
R.A. Cole, M. Fanty, M. Gopalakrishnan, R.D.T. Janssen, Speaker-independent name retrieval from spellings using a database of 50000 names international conference on acoustics, speech, and signal processing. pp. 325- 328 ,(1991) , 10.1109/ICASSP.1991.150342
Michael T. Rossides, Voice input system for data retrieval Journal of the Acoustical Society of America. ,vol. 99, pp. 3284- ,(1993) , 10.1121/1.414982
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
F.K. Soong, E.-F. Huang, A tree-trellis based fast search for finding the N-best sentence hypotheses in continuous speech recognition international conference on acoustics, speech, and signal processing. pp. 705- 708 ,(1991) , 10.1109/ICASSP.1991.150437