A Neural Vocoder with Hierarchical Generation of Amplitude and Phase Spectra for Statistical Parametric Speech Synthesis.
Yang AiZhen-Hua LingPublished in: CoRR (2019)
Keyphrases
- speech synthesis
- speech recognition
- text to speech
- phase information
- neural network
- prosodic features
- vocal tract
- network architecture
- statistical analysis
- data driven
- statistical methods
- information theoretic
- hierarchical structure
- hierarchical clustering
- statistical models
- parametric models
- instantaneous frequency
- generation process
- confidence intervals
- kernel density estimators
- neural model
- hierarchical model
- training phase
- hidden markov models
- face recognition
- image processing
- information retrieval
- machine learning