Zero-Shot vs. Few-Shot Multi-Speaker TTS Using Pre-trained Czech SpeechT5 Model.

Jan Lehecka Zdenek Hanzlícek Jindrich Matousek Daniel Tihelka

Published in: CoRR (2024)

Keyphrases

pre trained
face recognition
probabilistic model
support vector
pairwise
viewpoint
supervised learning
video data
statistical model
parametric models