Login / Signup

Label-free Knowledge Distillation with Contrastive Loss for Light-weight Speaker Recognition.

Zhiyuan PengXuanji HeKe DingTan LeeGuanglu Wan
Published in: ISCSLP (2022)
Keyphrases
  • lightweight
  • speaker recognition
  • vector quantization
  • neural network
  • multiresolution
  • gaussian mixture model
  • face recognition
  • multiscale
  • prior knowledge
  • speaker verification