On the Use of Self-Supervised Speech Representations in Spontaneous Speech Synthesis.

Siyang Wang Gustav Eje Henter Joakim Gustafson Éva Székely

Published in: CoRR (2023)

Keyphrases

wavelet transform
subband
speech synthesis
speech recognition
text to speech
vocal tract
prosodic features
speech corpus
hidden markov models
neural network
speech signal
image processing
pattern recognition
natural language
higher level
multiple representations