Improving Speaker Verification for Reverberant Conditions with Deep Neural Network Dereverberation Processing

Peter Guzewich, Stephen A. Zahorian


We present an improved method for training Deep Neural Networks for dereverberation and show that it can improve performance for the speech processing tasks of speaker verification and speech enhancement. We replicate recently proposed methods for dereverberation using Deep Neural Networks and present our improved method, highlighting important aspects that influence performance. We then experimentally evaluate the capabilities and limitations of the method with respect to speech quality and speaker verification to show that ours achieves better performance than other proposed methods.


 DOI: 10.21437/Interspeech.2017-461

Cite as: Guzewich, P., Zahorian, S.A. (2017) Improving Speaker Verification for Reverberant Conditions with Deep Neural Network Dereverberation Processing. Proc. Interspeech 2017, 171-175, DOI: 10.21437/Interspeech.2017-461.


@inproceedings{Guzewich2017,
  author={Peter Guzewich and Stephen A. Zahorian},
  title={Improving Speaker Verification for Reverberant Conditions with Deep Neural Network Dereverberation Processing},
  year=2017,
  booktitle={Proc. Interspeech 2017},
  pages={171--175},
  doi={10.21437/Interspeech.2017-461},
  url={http://dx.doi.org/10.21437/Interspeech.2017-461}
}