ISCA Archive Interspeech 2008
ISCA Archive Interspeech 2008

Unsupervised re-scoring of observation probability based on maximum entropy criterion by using confidence measure with telephone speech

Carlos Molina, Nestor Becerra Yoma, Fernando Huenupan, Claudio Garreton

This paper describes a two-step Viterbi decoding based on reinforcement learning and information theory with telephone speech. The idea is to strength or weaken HMM's by using Bayes-based confidence measure (BBCM) and distances between models. If HMM's in the N-best list show a low BBCM, the second Viterbi decoding will prioritize the search on neighboring models according to their distances to the N-best HMM's. The current reinforcement learning mechanism is modeled as the linear combination of two metrics or information sources. Moreover, a criterion based on incremental conditional entropy maximization to optimize a linear combination of metrics or information sources is also presented. As shown here, the method requires only one adapting utterance and can lead to a reduction in WER as high as 10.9%.


doi: 10.21437/Interspeech.2008-295

Cite as: Molina, C., Yoma, N.B., Huenupan, F., Garreton, C. (2008) Unsupervised re-scoring of observation probability based on maximum entropy criterion by using confidence measure with telephone speech. Proc. Interspeech 2008, 1016-1019, doi: 10.21437/Interspeech.2008-295

@inproceedings{molina08_interspeech,
  author={Carlos Molina and Nestor Becerra Yoma and Fernando Huenupan and Claudio Garreton},
  title={{Unsupervised re-scoring of observation probability based on maximum entropy criterion by using confidence measure with telephone speech}},
  year=2008,
  booktitle={Proc. Interspeech 2008},
  pages={1016--1019},
  doi={10.21437/Interspeech.2008-295}
}