Deep speech inpainting of time-frequency masks.
Mikolaj KeglerPierre BeckmannMilos CernakPublished in: CoRR (2019)
Keyphrases
- speech recognition
- image restoration
- speech signal
- wavelet transform
- signal processing
- audio visual
- denoising
- edge detection
- frequency domain
- text to speech
- endpoint detection
- automatic speech recognition
- automatic speech recognition systems
- recognition engine
- speech processing
- image inpainting
- spoken language
- language acquisition
- image processing
- texture synthesis
- multi modal
- broadcast news
- speaker verification
- speech synthesis
- resolution enhancement
- vocal tract
- fourier transform
- multiscale