ISCA Archive Interspeech 2005
ISCA Archive Interspeech 2005

Using output probability distribution for improving speech recognition in adverse environment

Shilei Huang, Xiang Xie, Jingming Kuang

This paper proposed a method to improve the accuracy of small vocabulary isolated word speaker-independent speech recognition in adverse environment. The proposed approach is implemented by using Output Probability Distributions (OPDs) and Support Vector Machine (SVM). OPDs improve the system performance by modeling inter-word relationships; then SVM classifiers are used to discriminate the difference between OPD models. The system was tested using isolated Mandarin digits database, corrupted with the NOISEX-92 database. The experiments have achieved good result in noise conditions, the WER dropped about 30% on average when compared to the HMM recognizer.


doi: 10.21437/Interspeech.2005-221

Cite as: Huang, S., Xie, X., Kuang, J. (2005) Using output probability distribution for improving speech recognition in adverse environment. Proc. Interspeech 2005, 929-932, doi: 10.21437/Interspeech.2005-221

@inproceedings{huang05b_interspeech,
  author={Shilei Huang and Xiang Xie and Jingming Kuang},
  title={{Using output probability distribution for improving speech recognition in adverse environment}},
  year=2005,
  booktitle={Proc. Interspeech 2005},
  pages={929--932},
  doi={10.21437/Interspeech.2005-221}
}