Maximum likelihood multiple subspace projections for hidden Markov models

  title={Maximum likelihood multiple subspace projections for hidden Markov models},
  author={Mark J. F. Gales},
  journal={IEEE Trans. Speech and Audio Processing},
The first stage in many pattern recognition tasks is to generate a good set of features from the observed data. Usually, only a single feature space is used. However, in some complex pattern recognition tasks the choice of a good feature space may vary depending on the signal content. An example is in speech recognition where phone dependent feature subspaces may be useful. Handling multiple subspaces while still maintaining meaningful likelihood comparisons between classes is a key issue. This… CONTINUE READING
Highly Cited
This paper has 51 citations. REVIEW CITATIONS
35 Citations
25 References
Similar Papers


Publications citing this paper.
Showing 1-10 of 35 extracted citations

52 Citations

Citations per Year
Semantic Scholar estimates that this publication has 52 citations based on the available data.

See our FAQ for additional information.


Publications referenced by this paper.
Showing 1-10 of 25 references

Investigation of silicon-auditory models and generalization of linear discriminant analysis for improved speech recognition,

  • N. Kumar
  • 1997
Highly Influential
10 Excerpts

The acoustic-modeling problem in automatic speech recognition,

  • P. Brown
  • Ph.D. dissertation, IBM T. J. Watson Res
  • 1987
Highly Influential
5 Excerpts

ovey, “Very largescale MMIE training for conversational telephone speech recognition,

  • P. C. Woodland, P D.
  • respectively. He was a Consultant with Roke Manor…
  • 2000
1 Excerpt

Similar Papers

Loading similar papers…