AeGAN: Time-Frequency Speech Denoising via Generative Adversarial Networks.
Sherif AbdulatifKarim ArmaniousKarim GuirguisJayasankar T. SajeevBin YangPublished in: EUSIPCO (2020)
Keyphrases
- denoising
- wavelet packet
- image denoising
- frequency domain
- social networks
- noise removal
- noisy images
- total variation
- network structure
- network design
- natural images
- unsupervised learning
- network analysis
- image processing
- recognition engine
- speech signal
- speech recognition
- generative model
- data driven
- multi agent
- computer networks
- wavelet domain
- non stationary
- audio visual
- language model
- discriminative learning
- signal analysis
- text to speech
- speech synthesis
- multiscale