Generative adversarial network-based postfilter for statistical parametric speech synthesis.
Takuhiro KanekoHirokazu KameokaNobukatsu HojoYusuke IjimaKaoru HiramatsuKunio KashinoPublished in: ICASSP (2017)
Keyphrases
- speech synthesis
- speech recognition
- text to speech
- vocal tract
- data driven
- generative model
- kernel density estimators
- prosodic features
- statistical information
- statistical approaches
- statistical models
- image coding
- lapped transforms
- statistical methods
- data sets
- statistical analysis
- pattern recognition
- unsupervised learning
- multi agent
- similarity measure
- image processing
- feature selection