… U-Net with a gated recurrent unit and an efficient channel attention mechanism for real-time
single-channel speech … Additionally, the U-Net architecture uses gated recurrent units (GRU), …

… variational U-Net architecture over its classic, non-variational version in signal enhancement
… with the variational U-Net compared to the recurrent mask estimation network baseline. …

… The recurrent U-net architecture was extended in [11], where recurrent layers were added
… In the present work, we propose a U-net-based neural network for speech enhancement. In …

… Recently speech enhancement auto-encoders with recurrent … In this work, several architectures
were tested, including: (1) a pipeline architecture of LSTM-based speech enhancement …

… In this work, we propose a new U-Net-based speech enhancement model using a novel
lightweight and e±cient Shu®le Attention (SA), Gated Recurrent Unit (GRU), … Model architecture …

… -based methods, such as the deep recurrent neural network (DRNN), SE generative … based
on a multi-task learning U-Net (MTU-Net) architecture to provide a better estimate of the IRM …

… is the proposal of four simple enhancements to the Wave-U-Net architecture that contribute
to the … and recognition using multi-task learning of long short-term memory recurrent neural …

… use the LRS2BBC dataset [25] to learn a model that can work robustly in real-world environments.
Moreover, we extend the Unet using the proposed Recurrent… our model architecture in …

… In this work, therefore, a time-domain U-Net speech enhancement model which … These
methods include convolutional neural network (CNN), recurrent neural … Model Architecture …

… networks (CNNs) and recurrent neural networks (RNNs), … of speech, we here propose a
context-aware U-Net based on … (TSTM) into the U-Net architecture. In our proposed model, the …