Login / Signup

fairseq S\^2: A Scalable and Integrable Speech Synthesis Toolkit.

Changhan WangWei-Ning HsuYossi AdiAdam PolyakAnn LeePeng-Jen ChenJiatao GuJuan Pino
Published in: EMNLP (Demos) (2021)
Keyphrases
  • speech synthesis
  • speech recognition
  • text to speech
  • prosodic features
  • highly scalable
  • vocal tract
  • speech corpus
  • bayesian networks
  • hidden markov models
  • lightweight
  • web scale
  • memory efficient
  • gradient field