The DKU System for the Speaker Recognition Task of the 2019 VOiCES from a Distance Challenge

Danwei Cai, Xiaoyi Qin, Weicheng Cai, Ming Li


In this paper, we present the DKU system for the speaker recognition task of the VOiCES from a distance challenge 2019. We investigate the whole system pipeline for the far-field speaker verification, including data pre-processing, short-term spectral feature representation, utterance-level speaker modeling, backend scoring, and score normalization. Our best single system employs a residual neural network trained with angular softmax loss. Also, the weighted prediction error algorithms can further improve performance. It achieves 0.3668 minDCF and 5.58% EER on the evaluation set by using a simple cosine similarity scoring. Finally, the submitted primary system obtains 0.3532 minDCF and 4.96% EER on the evaluation set.


 DOI: 10.21437/Interspeech.2019-1435

Cite as: Cai, D., Qin, X., Cai, W., Li, M. (2019) The DKU System for the Speaker Recognition Task of the 2019 VOiCES from a Distance Challenge. Proc. Interspeech 2019, 2493-2497, DOI: 10.21437/Interspeech.2019-1435.


@inproceedings{Cai2019,
  author={Danwei Cai and Xiaoyi Qin and Weicheng Cai and Ming Li},
  title={{The DKU System for the Speaker Recognition Task of the 2019 VOiCES from a Distance Challenge}},
  year=2019,
  booktitle={Proc. Interspeech 2019},
  pages={2493--2497},
  doi={10.21437/Interspeech.2019-1435},
  url={http://dx.doi.org/10.21437/Interspeech.2019-1435}
}