Login / Signup

Using Multiple Reference Audios and Style Embedding Constraints for Speech Synthesis.

Cheng GongLongbiao WangZhenhua LingJu ZhangJianwu Dang
Published in: ICASSP (2022)
Keyphrases
  • speech synthesis
  • speech recognition
  • constraint satisfaction
  • combining multiple
  • constrained optimization
  • real time
  • neural network
  • nonlinear dimensionality reduction
  • text to speech