Rethinking complex-valued deep neural networks for monaural speech enhancement.
Haibin WuKe TanBuye XuAnurag KumarDaniel WongPublished in: CoRR (2023)
Keyphrases
- complex valued
- speech enhancement
- neural network
- activation function
- recurrent neural networks
- noise reduction
- noisy environments
- real valued
- signal to noise ratio
- single channel
- artificial neural networks
- back propagation
- feed forward
- speech signal
- linear prediction
- multi valued
- pattern recognition
- hidden layer
- fuzzy logic
- vocal tract
- multilayer perceptron
- linear equations
- wiener filter
- neural nets
- background noise
- edge detection
- genetic algorithm
- multi layer perceptron
- radial basis function
- signal processing