RHR-Net: A Residual Hourglass Recurrent Neural Network for Speech Enhancement.
Jalal AbdulbaqiYue GuIvan MarsicPublished in: CoRR (2019)
Keyphrases
- recurrent neural networks
- speech enhancement
- noisy environments
- noise reduction
- signal to noise ratio
- single channel
- neural network
- feed forward
- speech signal
- linear prediction
- reservoir computing
- artificial neural networks
- recurrent networks
- vocal tract
- complex valued
- wiener filter
- echo state networks
- multi channel
- background noise
- additive noise
- edge preserving
- smoothing algorithm
- prior information
- maximum likelihood
- edge detection
- image processing