A stochastic segment model for phoneme-based continuous speech recognition

  title={A stochastic segment model for phoneme-based continuous speech recognition},
  author={Mari Ostendorf and Salim Roukos},
  journal={IEEE Trans. Acoustics, Speech, and Signal Processing},
Developing accurate and robust phonetic models for the different speech sounds is a major challenge for high-performance continuous speech recognition. In this paper, we introduce a new approach to modeling variable-duration phonemes, called the stochastic segment model. A phoneme X i s observed as a variable-length sequence of frames where each f rame is represented by a parameter vector and where the length of the sequence is random. The stochastic segment model consists of 1) a time warping… CONTINUE READING
Highly Cited
This paper has 198 citations. REVIEW CITATIONS

From This Paper

Figures, tables, and topics from this paper.


Publications citing this paper.
Showing 1-10 of 121 extracted citations

Sequence Prediction with Neural Segmental Models

ArXiv • 2017
View 4 Excerpts
Highly Influenced

Analysis of User Similarity Measures Using GPS Trajectories

2017 International Conference on Computing, Communication, Control and Automation (ICCUBEA) • 2017

Continuous speech recognition system for chhattisgarhi

2017 International Conference on Communication and Signal Processing (ICCSP) • 2017

198 Citations

Citations per Year
Semantic Scholar estimates that this publication has 198 citations based on the available data.

See our FAQ for additional information.


Publications referenced by this paper.
Showing 1-6 of 6 references

1151 M

Acoust.. Speech, Signul Processing, New York, NY. Apr.
A. Bu\h and G . E. Kopec, "Network-based connected digit recTram. Acoust. • 1988


COJIJ. Acoust., Speech. Signul Processing. Tampa. FL. Mar.
1205-1208, Paper 31.3. Y. L. Chow, R. M. Schwartz. S . Roucos. 0. A. Kimball. P. J . Price. G. F. Kubala, M. 0 . Dunham. M. A. Krasner, and J . Makhoul. "The • 1985

Matrix quantizer design from LPC speech using the generalized Lloyd algorithm.

Tsao, R. M. Gray
Spredi. Signfil Proc?ssing. Paris. France. May • 1982

Stmri.sri~d Theory ~f Signul Dcrrc.tion

C. W. Heljtroni
Mari Ostendorf received the B.S.. M.S. . and Ph.D. degrees in 1980 • 1981

Bocchieri and G . Doddington , " Frame - specitic stati 5 tical feature 5 for speaker - independent speech recognition . " lEEE Trtrns . Acoir . \ r . . Speech , Signul Proc - essitig

R. Bakis L. R. Bahl, P. S. Cohen, +4 authors S. E. Levinson

Similar Papers

Loading similar papers…