Improving Prosody Modelling with Cross-Utterance BERT Embeddings for End-to-end Speech Synthesis.
Guanghui XuWei SongZhengchen ZhangChao ZhangXiaodong HeBowen ZhouPublished in: CoRR (2020)
Keyphrases
- speech synthesis
- end to end
- speech recognition
- text to speech
- prosodic features
- vocal tract
- automatic speech recognition
- pattern recognition
- hidden markov models
- wireless ad hoc networks
- ad hoc networks
- admission control
- language model
- congestion control
- content delivery
- speech signal
- text localization and recognition
- high bandwidth
- rate allocation
- real world
- application layer
- image processing
- information retrieval
- neural network