Linear discriminant analysis for improved large vocabulary continuous speech recognition
作者： R. Haeb-Umbach , H. Ney
DOI: 10.1109/ICASSP.1992.225984
关键词:
摘要: The interaction of linear discriminant analysis (LDA) and a modeling approach using continuous Laplacian mixture density HMM is studied experimentally. largest improvements in speech recognition could be obtained when the classes for LDA transform were defined to sub-phone units. On 12000 word German task with small overlap between training test vocabulary reduction error rate by one-fifth was achieved compared case without LDA. development set DARPA RM1 reduced one-third. For speaker-dependent no-grammar case, averaged over 12 speakers 9.9%. This recognizer only 47 Viterbi-trained context-independent phonemes. >
acm.org
sci-hub.se 参考文章(9)
Peter F. Brown, The acoustic-modeling problem in automatic speech recognition Interim Report Carnegie-Mellon Univ. ,(1987) , 10.21236/ADA188529
G.R. Doddington, Phonetically sensitive discriminants for improved speech recognition international conference on acoustics, speech, and signal processing. pp. 556- 559 ,(1989) , 10.1109/ICASSP.1989.266487
Keinosuke Fukunaga, Introduction to statistical pattern recognition (2nd ed.) Academic Press Professional, Inc.. ,(1990)
M.J. Hunt, S.M. Richardson, D.C. Bateman, A. Piau, An investigation of PLP and IMELDA acoustic representations and of their potential for combination international conference on acoustics, speech, and signal processing. pp. 881- 884 ,(1991) , 10.1109/ICASSP.1991.150480
S.A. Zahorian, D. Qian, A.J. Jagharghi, Acoustic-phonetic transformations for improved speaker-independent isolated word recognition international conference on acoustics, speech, and signal processing. pp. 561- 564 ,(1991) , 10.1109/ICASSP.1991.150401
L.C. Wood, D.J.B. Pearce, F. Novello, Improved vocabulary-independent sub-word HMM modelling [Proceedings] ICASSP 91: 1991 International Conference on Acoustics, Speech, and Signal Processing. pp. 181- 184 ,(1991) , 10.1109/ICASSP.1991.150307
Keinosuke Fukunaga, Introduction to Statistical Pattern Recognition ,(1972)
G. Yu, W. Russell, R. Schwartz, J. Makhoul, Discriminant analysis and supervised vector quantization for continuous speech recognition international conference on acoustics, speech, and signal processing. pp. 685- 688 ,(1990) , 10.1109/ICASSP.1990.115850
D.B. Paul, The Lincoln tied-mixture HMM continuous speech recognizer international conference on acoustics, speech, and signal processing. pp. 329- 332 ,(1991) , 10.1109/ICASSP.1991.150343