A novel static parameter calculation method for model compensation

Suliang Bu, Yunxin Zhao, Yanmin Qian, Kai Yu

Abstract

Vector Taylor Series (VTS) based model compensation approach has been successfully applied to various robust speech recognition tasks. In this paper, we propose a novel method of variable transformation to calculate the static statistics. In addition, we provide a detailed explanation of VTS and random variable transformations adopted in some recent papers. Experiments on Aurora 4 showed that the proposed approach obtained 22.8% relative WER reduction over the traditional first-order VTS methods.
BibTeX
@inproceedings{bu2015novel,
  author = {Suliang Bu and Yunxin Zhao and Yanmin Qian and Kai Yu},
  title = {{A novel static parameter calculation method for model compensation}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2015},
  doi = {10.1109/icassp.2015.7178824},
}

Loading related content...

Figure