English

XMUSPEECH System for VoxCeleb Speaker Recognition Challenge 2021

Audio and Speech Processing 2021-09-07 v1 Sound

Abstract

This paper describes the XMUSPEECH speaker recognition and diarisation systems for the VoxCeleb Speaker Recognition Challenge 2021. For track 2, we evaluate two systems including ResNet34-SE and ECAPA-TDNN. For track 4, an important part of our system is VAD module which greatly improves the performance. Our best submission on the track 4 obtained on the evaluation set DER 5.54% and JER 27.11%, while the performance on the development set is DER 2.92% and JER 20.84%.

Cite

@article{arxiv.2109.02549,
  title  = {XMUSPEECH System for VoxCeleb Speaker Recognition Challenge 2021},
  author = {Jie Wang and Fuchuang Tong and Zhicong Chen and Lin Li and Qingyang Hong and Haodong Zhou},
  journal= {arXiv preprint arXiv:2109.02549},
  year   = {2021}
}
R2 v1 2026-06-24T05:43:24.991Z