Improving End-to-End Speech Synthesis with Local Recurrent Neural Network Enhanced Transformer.
Yibin ZhengXinhui LiFenglong XieLi LuPublished in: ICASSP (2020)
Keyphrases
- end to end
- recurrent neural networks
- speech synthesis
- speech recognition
- text to speech
- complex valued
- neural network
- vocal tract
- recurrent networks
- feed forward
- echo state networks
- prosodic features
- multipath
- reservoir computing
- congestion control
- wireless ad hoc networks
- hidden layer
- fuzzy logic
- artificial neural networks
- admission control
- transport layer
- internet protocol
- scalable video
- ad hoc networks
- high bandwidth
- application layer
- rate allocation
- fault diagnosis
- hidden markov models
- expert systems