ISCA Archive Interspeech 2008
ISCA Archive Interspeech 2008

Using syllable nuclei locations to improve automatic speech recognition in the presence of burst noise

Chris D. Bartels, Jeff A. Bilmes

In this work we combine a conventional phone-based automatic speech recognizer with a classifier that detects syllable locations. This is done using a dynamic Bayesian network. Using oracle syllable detections we achieve a 17% relative reduction in word error rate on the 500 word task of the SVitchboard corpus. Using estimated locations we achieve a 2.1% relative reduction which is significant at the 0.02 level. The improvement in the estimated case is from reducing insertions caused by burst noise.


doi: 10.21437/Interspeech.2008-130

Cite as: Bartels, C.D., Bilmes, J.A. (2008) Using syllable nuclei locations to improve automatic speech recognition in the presence of burst noise. Proc. Interspeech 2008, 2406-2409, doi: 10.21437/Interspeech.2008-130

@inproceedings{bartels08_interspeech,
  author={Chris D. Bartels and Jeff A. Bilmes},
  title={{Using syllable nuclei locations to improve automatic speech recognition in the presence of burst noise}},
  year=2008,
  booktitle={Proc. Interspeech 2008},
  pages={2406--2409},
  doi={10.21437/Interspeech.2008-130}
}