HMM-based estimation of unreliable spectral components for noise robust speech recognition

@inproceedings{Borgstrom2008HMMbasedEO,
  title={HMM-based estimation of unreliable spectral components for noise robust speech recognition},
  author={Bengt J. Borgstrom and Abeer Alwan},
  booktitle={INTERSPEECH},
  year={2008}
}
This paper presents a novel approach for reconstructing unreliable spectral components, which utilizes HMM-based missing feature algorithms, and applies them to noise robust speech recognition. The proposed technique uses the forwardbackward algorithm to estimate corrupt spectrographic data based on nearby reliable features, noisy observations, and on an underlying statistical model. The estimation process can be applied based on intra-channel information, intra-feature information, or a… CONTINUE READING
6 Citations
10 References
Similar Papers

References

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

Acoustic and Environemntal Robustness in Automatic Speech Recognition

  • A. Acero
  • Kluwer
  • 1993
1 Excerpt

Similar Papers

Loading similar papers…