Multimodal speech synthesis architecture for unsupervised speaker adaptation.
Hieu-Thi LuongJunichi YamagishiPublished in: CoRR (2018)
Keyphrases
- speech synthesis
- speech recognition
- speaker adaptation
- vocal tract
- speech recognizer
- speaker independent
- text to speech
- prosodic features
- speaker dependent
- unsupervised learning
- multi modal
- automatic speech recognition
- pattern recognition
- speech signal
- language model
- semi supervised
- hidden markov models
- computer vision
- neural network
- probabilistic model
- machine learning