Diphone subspace mixture trajectory models for HMM complementation

Speech Communication

Volumn 38, Issue 3-4, 2002, Pages 237-265

  Reinhard, K ^a,b   Niranjan, M ^a  

^a UNIVERSITY OF CAMBRIDGE   (United Kingdom)

^b ERICSSON RESEARCH   (Germany)

Author keywords

Diphones; N best rescoring; PCA; Speech dynamics; Subspace trajectory; Time constraint

Indexed keywords

DATABASE SYSTEMS; ELECTROMAGNETIC WAVES; MARKOV PROCESSES; SPEECH COMMUNICATION; WAVE FILTERS;

DIPHONES; PHONE RECOGNITION;

SPEECH RECOGNITION;

EID: 0036836695     PISSN: 01676393     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0167-6393(01)00054-1     Document Type: Article

References (44)

1. Atal B. Efficient coding of LPC parameters by temporal decomposition. Internat. Conf. Acoust. Speech Signal Process. 1:1983;81-84.
2. Blackburn, C., 1996. Articulatory methods for speech production and recognition. Ph.D. Thesis, Engineering Department, Cambridge University.
3. Blackburn C., Young S. Towards improved speech recognition using a speech production model. Eur. Conf. Speech Commun. Technol. (Eurospeech). 3:1995;1623-1626.
4. Bourlard H. Towards increasing speech recognition error rates. Eur. Conf. Speech Commun. Technol. (Eurospeech). 2:1995;883-894.
5. Bridle, J., Deng, L., Picone, J., Richards, H., Ma, J., Kamm, T., Schuster, M., Pike, S., Regan, R., 1998. An investigation of segmental hidden dynamic models of speech coarticulation for automatic speech recognition. In: Workshop on Language Engineering, Johns Hopkins University.
6. Deng L. A dynamic, feature-based approach to the interface between phonology and phonetics for speech modeling and recognition. Speech Communication. 24:1998;299-323.
7. Deng L., Aksmanovic M. Speaker-indepedent phonetic classification using hidden Markov models with mixture of trend functions. IEEE Trans. Speech Audio Process. 5(4):1997;319-324.
8. Deng L., Aksmanovic M., Sun X., Wu C. Speech recognition using hidden Markov models with polynomial regression functions as nonstationary states. IEEE Trans. Speech Audio Process. 2(4):1994;507-520.
9. Fukada T., Sagisaka Y., Paliwal K. Model parameter estimation for mixture density polynomial segment models. Internat. Conf. Acoust. Speech Signal Process. 2:1997;1403-1406.
10. Furui S. Speaker independent isolated word recognition using dynamic features of speech spectrum. IEEE Trans. Acoust. Speech Signal Process. 34(1):1986;52-59.
11. Gales M., Young S. Segmental hidden Markov models. Eur. Conf. Speech Commun. Technol. (Eurospeech). 3:1993;1579-1582.
12. Garofolo, J., 1988. Getting started with the DARPA TIMIT CD-ROM: an acoustic phonetic continuous speech database. Technical report, National Institute of Standards and Technology (NIST).
13. Ghitza O., Sondhi M. Hidden Markov models with templates as non-stationary states: an application to speech recognition. Comput. Speech Language. 2:1993;101-119.
14. Gillick L., Cox S. Some statistical issues in the comparison of speech recognition algorithms. Internat. Conf. Acoust. Speech Signal Process. 1:1989;532-535.
15. Gish H., Ng K. A segmental speech model with applications to word spotting. Internat. Conf. Acoust. Speech Signal Process. 2:1993;447-450.
16. Gish H., Ng K. Parametric trajectory models for speech recognition. Internat. Conf. Spoken Language Process. 1:1996;466-469.
17. Goldenthal, W., 1994. Statistical trajectory models for phonetic recognition. Ph.D. Thesis, Department of Aeronautics and Astronautics, MIT.
18. Hermansky H., Sharma S. Temporal patterns (TRAPS) in ASR of noisy speech. Internat. Conf. Acoust. Speech Signal Process. 1:1999;289-292.
19. Holmes W., Russell M. Speech recognition using a linear dynamic segmental HMM. Eur. Conf. Speech Commun. Technol. (Eurospeech). 3:1995;1611-1614.
20. Holmes W., Russell M. Linear dynamic segmental HMMs: Variability representation and training procedure. Internat. Conf. Acoust. Speech Signal Process. 2:1997;1399-1402.
21. Hon H.-W., Wang K. Unified frame and segment based models for automatic speech recognition. Internat. Conf. Acoust. Speech Signal Process. 2:2000;1017-1020.
22. Hosom J.-P., Cole R. A diphone-based digit recognition system using neural networks. Internat. Conf. Acoust. Speech Signal Process. 4:1997;3369-3372.
23. Kannan, A., Ostendorf, M., Rohlicek, J., 1992. Weight estimation for N-best rescoring. In: Proc. DARPA Speech and Natural Language Workshop, pp. 455-456.
24. Lamel, L., Kassel, R., Seneff, S., 1986. Speech database development: design and analysis of the acoustic-phonetic corpus. In: Proc. Speech Recognition Workshop (DARPA), pp. 100-109.
25. Lloyd S. Least squares quantization in PCM. IEEE Trans. Inf. Theory. 28(2):1982;129-137.
26. Marcus, S., van Lieshout, R., 1984. Temporal decomposition of speech. Annual Progress Report, IPO, Eindhoven, pp. 25-31.
27. Marteau P., Bailly G., Janot-Giorgetti M. Stochastic model of diphone-like segments based on trajectory concepts. Internat. Conf. Acoust. Speech Signal Process. 1:1988;615-618.
28. Morgan N., Bourlard H., Greenberg S., Hermansky H. Stochastic perceptual auditory-events-based models for speech recognition. Internat. Conf. Spoken Language Process. 4:1994;1943-1946.
29. Morgan N., Bourlard H., Greenberg S., Hermansky H., Wu S. Stochastic perceptual models of speech. Inernat. Conf. Acoust. Speech Signal Process. 1:1995;397-400.
30. Niranjan M., Fallside F. On modelling the dynamics of speech patterns. Eur. Conf. Speech Technol. 1:1987;71-74.
31. Oja E. Subspace Methods of Pattern Recognition. 1983;Research Studies Press, Letchworth, UK.
32. O'Shaughnessy D. Speech Communication - Human and Machine. second ed. 2000;IEEE Press, New York.
33. Ostendorf M., Roukos S. A stochastic segment model for phoneme-based continuous speech recognition. IEEE Trans. Acoust. Speech Signal Process. 37(12):1989;1857-1869.
34. Ostendorf, M., Kannan, A., Austin, S., Kimball, O., Schwartz, R., Rohlicek, J., 1991. Integration of diverse recognition methodologies through reevaluation of N-best sentence hypotheses. In: Proc. DARPA Speech and Natural Language Workshop, pp. 83-87.
35. Ostendorf M., Digalakis V., Kimball O. From HMM's to segment models: A unified view of stochastic modeling for speech recognition. IEEE Trans. Speech Audio Process. 4(5):1996;360-378.
36. Picone J., Pike S., Regan R., Kamm T., Bridle J., Deng L., Ma Z., Richards H., Schuster M. Initial evaluation of hidden dynamic models on conversational speech. Internat. Conf. Acoust. Speech Signal Process. 1:1999;109-112.
37. Reinhard K., Niranjan M. Parametric subspace modeling of speech transitions. Speech Communication. 27(1):1999;19-42.
38. Richards H., Bridle J. The HDM: A segmental hidden dynamic model of coarticulation. Internat. Conf. Acoust. Speech Signal Process. 1:1999;357-360.
39. Russell M. A segmental HMM for speech pattern modelling. Internat. Conf. Acoust. Speech Signal Process. 2:1993;499-502.
40. Saul L., Rahim M. Markov processes on curves for automatic speech recognition. Adv. Neural Inf. Process. Syst. 11:1998;751-757.
41. Schwartz R., Chow Y.-L. The N-best algorithm: An efficient and exact procedure for finding the N most likely sentence hypotheses. Internat. Conf. Acoust. Speech Signal Process. 1:1990;81-84.
42. Schwartz R., Austin S., Kubala F., Makhoul J., Nguyen L., Placeway P., Zavaliagkos G. New uses for the N-best sentence hypotheses within the BYBLOS speech recognition system. Internat. Conf. Acoust. Speech Signal Process. 1:1992;1-4.
43. Viterbi A. Error bounds for convolutional codes and an asymptotically optimum decoding algorithm. IEEE Trans. Inf. Theory. IT-13(2):1967;260-269.
44. Young, S., Odell, J., Woodland, P., 1994. Tree-based state tying for high accuracy acoustic modeling. In: ARPA Workshop on Human Language Technology, pp. 307-312.