Detail publikačního výsledku

BUT System Description to VoxCeleb Speaker Recognition Challenge 2019

ZEINALI, H.; WANG, S.; SILNOVA, A.; MATĚJKA, P.; PLCHOT, O.

Original Title

BUT System Description to VoxCeleb Speaker Recognition Challenge 2019

English Title

BUT System Description to VoxCeleb Speaker Recognition Challenge 2019

Type

Paper in proceedings outside WoS and Scopus

Original Abstract

In this report, we describe the submission of Brno Universityof Technology (BUT) team to the VoxCeleb Speaker RecognitionChallenge (VoxSRC) 2019. We also provide a brief analysisof different systems on VoxCeleb-1 test sets. Submittedsystems for both Fixed and Open conditions are a fusion of4 Convolutional Neural Network (CNN) topologies. The firstand second networks have ResNet34 topology and use twodimensionalCNNs. The last two networks are one-dimensionalCNN and are based on the x-vector extraction topology. Someof the networks are fine-tuned using additive margin angularsoftmax. Kaldi FBanks and Kaldi PLPs were used as features.The difference between Fixed and Open systems lies in the usedtraining data and fusion strategy. The best systems for Fixedand Open conditions achieved 1.42 % and 1.26 % ERR on thechallenge evaluation set respectively.

English abstract

In this report, we describe the submission of Brno Universityof Technology (BUT) team to the VoxCeleb Speaker RecognitionChallenge (VoxSRC) 2019. We also provide a brief analysisof different systems on VoxCeleb-1 test sets. Submittedsystems for both Fixed and Open conditions are a fusion of4 Convolutional Neural Network (CNN) topologies. The firstand second networks have ResNet34 topology and use twodimensionalCNNs. The last two networks are one-dimensionalCNN and are based on the x-vector extraction topology. Someof the networks are fine-tuned using additive margin angularsoftmax. Kaldi FBanks and Kaldi PLPs were used as features.The difference between Fixed and Open systems lies in the usedtraining data and fusion strategy. The best systems for Fixedand Open conditions achieved 1.42 % and 1.26 % ERR on thechallenge evaluation set respectively.

Keywords

VoxCeleb Speaker Recognition Challenge, Deep NeuralNetworks, ResNet, x-vector, PLDA, Cosine distance

Key words in English

VoxCeleb Speaker Recognition Challenge, Deep NeuralNetworks, ResNet, x-vector, PLDA, Cosine distance

Authors

ZEINALI, H.; WANG, S.; SILNOVA, A.; MATĚJKA, P.; PLCHOT, O.

Released

14.09.2019

Location

Graz

Book

Proceedings of The VoxCeleb Challange Workshop 2019

Pages from

1

Pages to

4

Pages count

4

URL

https://arxiv.org/abs/1910.12592

Full text in the Digital Library

http://hdl.handle.net/

BibTex

@inproceedings{BUT168476,
  author="Hossein {Zeinali} and Shuai {Wang} and Anna {Silnova} and Pavel {Matějka} and Oldřich {Plchot}",
  title="BUT System Description to VoxCeleb Speaker Recognition Challenge 2019",
  booktitle="Proceedings of The VoxCeleb Challange Workshop 2019",
  year="2019",
  pages="1--4",
  address="Graz",
  url="https://arxiv.org/abs/1910.12592"
}

Documents

zeinali_Voxceleb_2019_Interspeech_Challenge_1_