Text-to-Speech Synthesis Using STFT Spectra Based on Low-/Multi-Resolution Generative Adversarial Networks.
Yuki SaitoShinnosuke TakamichiHiroshi SaruwatariPublished in: ICASSP (2018)
Keyphrases
- multiresolution
- text to speech synthesis
- generative model
- social networks
- text to speech
- data sets
- coarse to fine
- complex networks
- multi agent
- wavelet transform
- data mining
- hierarchical representation
- discriminative learning
- wavelet domain
- network structure
- complex systems
- data driven
- signal processing
- principal component analysis
- information retrieval