Self-attention Based Prosodic Boundary Prediction for Chinese Speech Synthesis.
Chunhui LuPengyuan ZhangYonghong YanPublished in: ICASSP (2019)
Keyphrases
- speech synthesis
- prosodic features
- speech recognition
- text to speech
- vocal tract
- prediction accuracy
- prediction algorithm
- text to speech synthesis
- speech corpus
- image processing
- multiscale
- hidden markov models
- prediction model
- speech signal
- linear prediction
- prediction error
- automatic speech recognition
- neural network
- noisy environments
- focus of attention
- object boundaries
- active contours
- language model
- information retrieval
- machine learning