Maximum mutual information based reduction strategies for cross-correlation based joint distributional modeling

@inproceedings{Bilmes1998MaximumMI,
  title={Maximum mutual information based reduction strategies for cross-correlation based joint distributional modeling},
  author={Jeff A. Bilmes},
  booktitle={ICASSP},
  year={1998}
}
In maximum-likelihood based speech recognition systems, it is important to accurately estimate the joint distribution of feature vectors given a particular acoustic model. In previous work, we showed we can boost accuracy in this task by modeling the joint distribution of time-localized feature vectors along with statistics relating those feature vectors to their surrounding context. In this work, we evaluate information preserving reduction strategies for those statistics. We claim that those… CONTINUE READING
Highly Cited
This paper has 52 citations. REVIEW CITATIONS

Citations

Publications citing this paper.
Showing 1-10 of 33 extracted citations

52 Citations

05'98'02'07'12'17
Citations per Year
Semantic Scholar estimates that this publication has 52 citations based on the available data.

See our FAQ for additional information.

References

Publications referenced by this paper.
Showing 1-7 of 7 references

Bilmes . Joint distributional modeling with cross - correlation based features

  • A. Jeff
  • 1997

An information-theoretical study of speech processing in the peripheral auditory system and cochlear nucleas

  • A. C. Morris
  • PhD thesis, Institut National Polytechnique De…
  • 1992
1 Excerpt

Similar Papers

Loading similar papers…