Correcting Misproducted Speech using Spectrogram Inpainting.
Talia Ben SimonFelix KreukFaten AwwadJacob T. CohenJoseph KeshetPublished in: CoRR (2022)
Keyphrases
- speech signal
- automatic speech recognition
- speech recognition
- noisy environments
- vocal tract
- speaker recognition
- image restoration
- speaker identification
- data sets
- non stationary
- spectral analysis
- resolution enhancement
- spoken dialogue systems
- image inpainting
- wigner distribution
- texture synthesis
- denoising
- automatic speech recognition systems
- energy distribution
- audio stream
- hearing impaired
- language acquisition
- speech synthesis
- spoken language
- emotion recognition
- audio visual
- hidden markov models
- video sequences