High-quality Speech Synthesis Using Super-resolution Mel-Spectrogram.
Leyuan ShengDong-Yan HuangEvgeniy N. PavlovskiyPublished in: CoRR (2019)
Keyphrases
- super resolution
- speech synthesis
- high quality
- speech recognition
- high resolution
- low resolution
- text to speech
- speech signal
- image reconstruction
- automatic speech recognition
- image restoration
- low resolution images
- high resolution images
- image super resolution
- depth map
- super resolution reconstruction
- motion blur
- image resolution
- low quality
- resolution enhancement
- super resolved
- video super resolution
- light field
- image patches
- multi frame
- higher quality
- higher resolution
- adaptive regularization
- single frame
- single image super resolution
- image quality
- language model
- hidden markov models
- high resolution video
- point correspondences
- high frequency
- remote sensing
- motion estimation
- pattern recognition