Speech recognition using hidden Markov models based on segmental statistics

Systems and Computers in Japan

Volumn 28, Issue 7, 1997, Pages 31-38

  Nakagawa, Seiichi ^a   Yamamoto, Kazumasa ^a  

^a TOYOHASHI UNIVERSITY OF TECHNOLOGY   (Japan)

Author keywords

Hidden Markov model (HMM); Segmental statistics; Speech recognition; Time variant feature

Indexed keywords

MARKOV PROCESSES; MATHEMATICAL MODELS; MATRIX ALGEBRA; REGRESSION ANALYSIS; VECTORS;

HIDDEN MARKOV MODELS (HMM);

SPEECH RECOGNITION;

EID: 0031153130     PISSN: 08821666     EISSN: None     Source Type: Journal    
DOI: 10.1002/(sici)1520-684x(19970630)28:7<31::aid-scj4>3.0.co;2-k     Document Type: Article

References (31)

1. S. Nakagawa, Y. Hirata, and Y. Hashimoto. Japanese phoneme recognition using continuous parameter hidden Markov models. J. Acoust. Soc. Japan, 46, No. 6, pp. 486-496 (June 1990). (in Japanese)
2. V. N. Gupta, M. Lennig, and P. Mermelstein. Integration of acoustic information in a large vocabulary word recognizer, ICASSP-87, II, pp. 697-700 (April).
3. E. Tsuboka. Hidden Markov model embedded dynamic features of speech spectrum. Tech. Rep. I.E.I.C.E., SP87-97. (in Japanese).
4. E. Tsuboka, Y. Takada, and H. Wakita. Neural predictive hidden Markov model. ICSLP-90, pp. 1341-1344.
5. L. Deng, M. Aksmanoric, X. Sun, and C.F.J. Wu. Speech recognition using hidden Markov models with polynomial regression functions as nonstationary states. IEEE Trans. Speech Audio Process., 2, No. 4, pp. 507-520 (April 1994).
6. Y. Hirata, I. Hayakawa, Y. Ono, and S. Nakagawa. Speech recognition by hidden Markov model using segmental statistics. Tech. Rep. I.E.I.C.E., SP90-69 (1990). (in Japanese)
7. S. Nakagawa, Y. Hirata, and Y. Ono. Syllable recognition by hidden Markov models using fixed-length segmental statistics. Trans. Inst. Electron Inf. Commun., J75-D-II. No. 5, pp. 843-851 (May 1992). (in Japanese)
8. C. J. Wellekens. Explicit correlation in hidden Markov model for speech recognition. ICASSP-87, I. pp. 383-386 (April).
9. F. J. Smith, J. Ming, P. O'Boyle, and A. D. Irvine. A hidden Markov model with optimized inter-frame dependence. ICASSP-95, I, pp. 209-212 (May).
10. K. Aikawa, H. Singer, H. Kawahara, and Y. Tohkura, A dynamic cepstrum incorporating time-frequency masking and its application to continuous speech recognition. ICASSP-93 II, pp. 668-671.
11. S. Nakagawa and Y. Enomoto. Speaker-independent phoneme and word recognition by statistical classification methods for time-sequential patterns. Trans. Inst. Electron Inf. Commun., J71-D, No. 10, pp. 1977-1983 (Oct. 1988). (in Japanese)
12. K. Ide, S. Makino, and K. Kido. Recognition of unvoiced plosives using time spectrum pattern. J. Acoust. Soc. Japan, 39, No. 5, pp. 321-329 (May 1983). (in Japanese)
13. M. Ostendorf and S. Roukos. A stochastic segment model for phoneme-based continuous speech recognition. IEEE Trans. Acoust. Speech Signal Process. 37, No. 12, pp. 1857-1869 (Dec. 1989).
14. M. Franzini and K. E Lee. Connectionist Viterbi training: A new hybrid for continuous speech recognition. ICASSP-90, I, pp. 425-428 (April).
15. S. Renal et al. Connectionist probability estimators in HMM speech recognition. IEEE Trans. Speech Audio Process., 2, No. 1, pp. 161-174 (Jan. 1994).
16. L. R. Bahl et al. Performance of the IBM large vocabulary continuous speech recognition system on the ARPA Wall Street Journal Task. ICASSP-95, I, pp. 41-44 (May).
17. J. He and H. Leich. A unified way of incorporating segmental features and segmental models into HMM. ICASSP-95, I, pp. 532-535 (May).
18. W. J. Holmes and M. J. Russell. Speech recognition using a linear dynamic segmental HMM. EUROSPEECH'95. III, pp. 1611-1614 (Sept.).
19. J. Ming, P. O'Boyle, and J. Smith. An HMM with optimized segment-dependent observations for speech recognition. EUROSPEECH'95, II, pp. 1475-1478 (Sept.).
20. M. Tachimori, H. Kanazawa, and H. Tsuboi. HMM based speech recognition using feature extraction by multiple coordinate systems. ASJ spring meeting, No. 2-8-1, pp. 95-96 (March 1996). (in Japanese)
21. E. L. Bocchieri and G. R. Doddington. Speaker-independent digit recognition with reference frame-specific distance measures. ICASSP-86, pp. 2669-2672 (April).
22. P. F. Brown. The acoustic-modeling problem in automatic speech recognition. Ph.D. thesis, CarnegieMellon University (1987).
23. L. R. Bahl, P. F. Brown, P. V. de Souza, and R. L. Mercer. Speech recognition with continuous-parameter hidden Markov models. ICASSP-88, I, pp. 40-43 (April).
24. K. Yamamoto and S. Nakagawa. Comparative evaluation of segmental unit input HMM and conditional density HMM. EUROSPEECH'95, III, pp. 1615-1618 (Sept.).
25. J. Ming and F. J. Smith. Isolated word recognition using interframe dependent hidden Markov models. IEEE Signal Process. Lett., 1, No. 12. pp. 188-190 (Dec. 1994).
26. S. Takahashi, T. Matsuoka, Y. Minami and K. Shikano. Phoneme HMMs constrained by frame correlations. ICASSP-93, II, pp. 219-222.
27. T. H. Applebaum and B. A. Hanson. Tradeoffs in the design of regression features for word recognition. EUROSPEECH'91, pp. 1203-1206 (Sept.).
28. S. Nakagawa and T. Koshikawa. Speaker adaptation for continuous parameter HMM by maximum a posteriori probability estimation. Tech. Rep. I.E.I.C.E., SP92-101 (1992). (in Japanese)
29. K. Yamamoto and S. Nakagawa. Evaluation of diagonal mixture distribution HMM with segmental unit inputting. ASJ autumn meeting, No. 2-3-1, pp. 49-50 (Sept. 1996). (in Japanese)
30. S. Nakagawa and K. Yamamoto. Evaluation of segmental unit input HMM. ICASSP-96, pp. 439-442.
31. Y. Tsurumi and S. Nakagawa. An unsupervised speaker adaptation method for continuous parameter HMM by maximum a posteriori probability estimation. 1CSLP-94, pp. 431-434 (Sept.).