The role of word-dependent coarticulatory effects in a phoneme-based speech recognition system
作者： Yen-Lu Chow , Richard Schwartz , Salim Roucos , Owen Kimball , Patti Price
DOI: 10.1109/ICASSP.1986.1168931
关键词:
摘要: This paper describes the results of our work in designing a system for large-vocabulary word recognition continuous speech. We generalize use context-dependent Hidden Markov Models (HMM) phonemes to take into account word-dependent coarticulatory effects, Robustness is assured by smoothing detailed models with less but more robust models. describe training and algorithms HMMs phonemes-in-context. On task 334-word vocabulary no grammar (i.e., branching factor 334), speaker-dependent mode, we show an average reduction error rate from 24% using context-independent phoneme models, 10% when
sci-hub.se 参考文章(3)
R. Schwartz, Y. Chow, S. Roucos, M. Krasner, J. Makhoul, Improved hidden Markov modeling of phonemes for continuous speech recognition international conference on acoustics, speech, and signal processing. ,vol. 9, pp. 21- 24 ,(1984) , 10.1109/ICASSP.1984.1172751
R. Schwartz, Y. Chow, O. Kimball, S. Roucos, M. Krasner, J. Makhoul, Context-dependent modeling for acoustic-phonetic recognition of continuous speech international conference on acoustics, speech, and signal processing. ,vol. 10, pp. 1205- 1208 ,(1985) , 10.1109/ICASSP.1985.1168283
J. Makhoul, S. Roucos, H. Gish, Vector quantization in speech coding Proceedings of the IEEE. ,vol. 73, pp. 1551- 1588 ,(1985) , 10.1109/PROC.1985.13340