Skip to main content
Springer Nature Link
Log in

HMM Parameter Adaptation Using the Truncated First-Order VTS and EM Algorithm for Robust Speech Recognition

  • Conference paper
Computational Intelligence and Security (CIS 2005)

Part of the book series: Lecture Notes in Computer Science ((LNAI,volume 3801))

Included in the following conference series:

Abstract

This paper presents a framework of HMM parameter adaptation technique for improving automatic speech recognition (ASR) performance in the noisy environments, which online combines the clean hidden Markov models (HMMs) with the noise model. Based on the given composite HMM corresponding to the initial recognition pass result and truncated vector Taylor series, the noise model in the cepstral domain is updated and refined using iterative Expectation-Maximization (EM) algorithm under maximum likelihood (ML) criterion. Experiments results show that the presented approach in this paper is found to greatly improve recognition performance under mismatched conditions.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Similar content being viewed by others

References

  1. Moreno, P.J., Raj, B., Stern, R.M.: A Vector Taylor Series Approach for Environment-Independent Speech Recognition. Proceedings of IEEE, 733–736 (1995)

    Google Scholar 

  2. Kim, N.S.: Statistical Linear Approximation for Environment Compensation. IEEE Signal Processing Letters 1, 8–10 (1998)

    Google Scholar 

  3. Shen, H., Liu, G., Guo, J., Li, Q.: Two-Domain Feature Compensation for Robust Speech Recognition. In: Wang, J., Liao, X.-F., Yi, Z. (eds.) ISNN 2005. LNCS, vol. 3497, pp. 351–356. Springer, Heidelberg (2005)

    Chapter  Google Scholar 

  4. Shen, H., Guo, J., Liu, G., Huang, P., Li, Q.: Environment Compensation Based on Maximum a Posteriori Estimation for Improved speech Recognition. In: Gelbukh, A., de Albornoz, Á., Terashima-Marín, H. (eds.) MICAI 2005. LNCS (LNAI), vol. 3789, pp. 854–862. Springer, Heidelberg (2005)

    Chapter  Google Scholar 

  5. Gales, M.J.F.: Model-Based Techniques for Noise Robust Speech Recognition. Ph.D. thesis, University of Cambridge (1995)

    Google Scholar 

  6. Acero, A., Deng, L., Kristjansson, K., Zhang, J.: HMM Adaptation Using Vector Taylor Series for Noisy Speech Recognition. In: Proceedings of ICSLP 2000 (2000)

    Google Scholar 

  7. Dempster, A.P., Laird, N.M., Rubin, D.B.: Maximum likelihood from incomplete data via the EM algorithm. J. of the Royal Statistical Society B 39, 1–38 (1977)

    MATH  MathSciNet  Google Scholar 

  8. Zu, Y.Q.: Issues in the Scientific Design of the Continuous Speech Database. Available, http://www.cass.net.cn/chinese/s18_yys/yuyin/report/report_1998.htm

  9. Varga, A., Steenneken, H.J.M., Tomilson, M., Jones, D.: The NOISEX–92 Study on the Effect of Additive Noise on Automatic Speech Recognition. Tech. Rep. DRA Speech Research Unit (1992)

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Beijing University of Posts and Telecommunications, Beijing, 100876, China

    Haifeng Shen, Jun Guo & Gang Liu

  2. University of Science and Technology Beijing, Beijing, 100083, China

    Qunxia Li

Authors
  1. Haifeng Shen
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Qunxia Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Jun Guo
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Gang Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

Editor information

Editors and Affiliations

  1. Microelectronic Instiute, Xidian University, 710071, Xi’an, China

    Yue Hao

  2. Department of Computer Science, Hong Kong Baptist University, Kowloon Tong, Hong Kong

    Jiming Liu

  3. School of Computer Science and Technology, Xidian University, Xi’an, China

    Yuping Wang

  4. Department of Computer Science, Hong Kong Baptist University, Hong Kong,  

    Yiu-ming Cheung

  5. School of Electrical and Electronic Engineering, University of Manchester, UK

    Hujun Yin

  6. Life Science Research Center, School of Electronic Engineering, Xidian University, 710071, Xi’an, Shaanxi, China

    Licheng Jiao

  7. Key Laboratory of Computer Networks and Information Security (Ministry of Education), Xidian University, 710071, Xi’an, China

    Jianfeng Ma

  8. National Laboratory of Antennas and Microwave Technology, Xidian University, 710071, Xi’an, Shanxi, P.R. China

    Yong-Chang Jiao

Rights and permissions

Reprints and permissions

Copyright information

© 2005 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Shen, H., Li, Q., Guo, J., Liu, G. (2005). HMM Parameter Adaptation Using the Truncated First-Order VTS and EM Algorithm for Robust Speech Recognition. In: Hao, Y., et al. Computational Intelligence and Security. CIS 2005. Lecture Notes in Computer Science(), vol 3801. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11596448_145

Download citation

  • DOI: https://doi.org/10.1007/11596448_145

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-30818-8

  • Online ISBN: 978-3-540-31599-5

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics