Joint and Adversarial Training with ASR for Expressive Speech Synthesis.
Kaili ZhangCheng GongWenhuan LuLongbiao WangJianguo WeiDawei LiuPublished in: ICASSP (2022)
Keyphrases
- speech synthesis
- speech recognition
- speech corpus
- automatic speech recognition
- vocal tract
- text to speech
- prosodic features
- training process
- online learning
- speech signal
- test set
- language model
- hidden markov models
- training set
- training algorithm
- database
- real time
- speech retrieval
- training phase
- training samples
- data mining
- noisy environments
- image acquisition
- small number
- multi agent
- e learning