Voice Cloning: Corentin's Improvisation On SV2TTS

By datasciencecentral - 2021-03-21

Description

Working with the audio production and engineering industry, I often wonder how the future of the voice talent market will look like with the assistance of art…

Summary

  • Corentin's Improvisation On SV2TTS Working with the audio production and engineering industry, I often wonder how the future of the voice talent market will look like with the assistance of artificial intelligence.
  • minwSLS(xij,S(eij,tij;wS)) He suggests training the synthesizer and vocoder separately.
  • When implemented, Corentin used the LibriSpeech dataset that he thought would give the best voice cloning similarity on unseen speakers.
  • Also, he used the Montreal Forced Aligner for Automatic Speech Recognition and to reduce background noises from synthesized spectrograms, the LogMMSE algorithm.

 

Topics

  1. NLP (0.22)
  2. UX (0.08)
  3. Backend (0.07)

Similar Articles

K-fold Cross Validation with PyTorch

By MachineCurve - 2021-02-02

Explanations and code examples showing you how to use K-fold Cross Validation for Machine Learning model evaluation/testing with PyTorch.