Skip to main content

Robust Model Adaptation Using Mean and Variance Transformations in Linear Spectral Domain

  • Conference paper
  • 1165 Accesses

Part of the Lecture Notes in Computer Science book series (LNISA,volume 3578)

Abstract

In this paper, we propose robust speech adaptation technique using continuous density hidden Markov models (HMMs) in unknown environments. This adaptation technique is an improved maximum likelihood linear spectral transformation (ML-LST) method, which aims to find appropriate noise parameters in the linear spectral domain. Previously, ML-LST and many transform-based adaptation algorithms have been applied to the Gaussian mean vectors of HMM systems. In the improved ML-LST for the rapid adaptation, the mean vectors and covariance matrices of an HMM based speech recognizer are transformed simultaneously using a small number of transformation parameters. It is shown that the variance transformation provides important information which can be used to handle environmental noise, in the similar manner that the mean transformation does.

Keywords

  • Discrete Cosine Transformation
  • Additive Noise
  • Adaptation Data
  • Speech Recognition System
  • Unknown Environment

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

This is a preview of subscription content, access via your institution.

Buying options

Chapter
USD   29.95
Price excludes VAT (USA)
  • DOI: 10.1007/11508069_20
  • Chapter length: 6 pages
  • Instant PDF download
  • Readable on all devices
  • Own it forever
  • Exclusive offer for individuals only
  • Tax calculation will be finalised during checkout
eBook
USD   109.00
Price excludes VAT (USA)
  • ISBN: 978-3-540-31693-0
  • Instant PDF download
  • Readable on all devices
  • Own it forever
  • Exclusive offer for individuals only
  • Tax calculation will be finalised during checkout
Softcover Book
USD   149.00
Price excludes VAT (USA)

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Leggetter, C.J., Woodland, P.C.: Maximum likelihood linear regression for speaker adaptation of continuous density hidden Markov models. Computer Speech and Language 9, 171–185 (1995)

    CrossRef  Google Scholar 

  2. Kim, D., Yook, D.: Fast channel adaptation for continuous density HMMs using maximum likelihood spectral transform. IEE Electronics Letters 40(10), 632–633 (2004)

    CrossRef  Google Scholar 

  3. Gales, M.J.F., Woodland, P.C.: Mean and variance adaptation within the MLLR framework. Computer Speech and Language 10, 249–264 (1996)

    CrossRef  Google Scholar 

  4. Gales, M.J.F.: Model-based techniques for noise robust speech recognition, Ph.D. thesis, Cambridge University (1995)

    Google Scholar 

  5. Baum, L.: An inequality and associated maximization technique in statistical estimation of probabilitic functions of a Markov process. Inequalities 3, 1–8 (1972)

    Google Scholar 

  6. Press, W., Teukolsky, S., Vetterling, W., Flannery, B.: Numerical recipes in C++, pp. 398–460. Cambridge University Press, Cambridge (2002)

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Editor information

Editors and Affiliations

Rights and permissions

Reprints and Permissions

Copyright information

© 2005 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Kim, D., Yook, D. (2005). Robust Model Adaptation Using Mean and Variance Transformations in Linear Spectral Domain. In: Gallagher, M., Hogan, J.P., Maire, F. (eds) Intelligent Data Engineering and Automated Learning - IDEAL 2005. IDEAL 2005. Lecture Notes in Computer Science, vol 3578. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11508069_20

Download citation

  • DOI: https://doi.org/10.1007/11508069_20

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-26972-4

  • Online ISBN: 978-3-540-31693-0

  • eBook Packages: Computer ScienceComputer Science (R0)