English

BUT VOiCES 2019 System Description

Audio and Speech Processing 2019-07-16 v1 Computation and Language Sound

Abstract

This is a description of our effort in VOiCES 2019 Speaker Recognition challenge. All systems in the fixed condition are based on the x-vector paradigm with different features and DNN topologies. The single best system reaches 1.2% EER and a fusion of 3 systems yields 1.0% EER, which is 15% relative improvement. The open condition allowed us to use external data which we did for the PLDA adaptation and achieved less than ~10% relative improvement. In the submission to open condition, we used 3 x-vector systems and also one i-vector based system.

Keywords

Cite

@article{arxiv.1907.06112,
  title  = {BUT VOiCES 2019 System Description},
  author = {Hossein Zeinali and Pavel Matějka and Ladislav Mošner and Oldřich Plchot and Anna Silnova and Ondřej Novotný and Ján Profant and Ondřej Glembek and Lukáš Burget},
  journal= {arXiv preprint arXiv:1907.06112},
  year   = {2019}
}
R2 v1 2026-06-23T10:20:20.651Z