Combining Multi-Perspective Attention Mechanism With Convolutional Networks for Monaural Speech Enhancement.
Tian LanYilan LyuWenzheng YeGuoqiang HuiZenglin XuQiao LiuPublished in: IEEE Access (2020)
Keyphrases
- multi perspective
- attention mechanism
- speech enhancement
- d scene
- visual attention
- speech signal
- signal to noise ratio
- noise reduction
- noisy environments
- speech recognition
- image segmentation
- higher order
- higher level
- subband
- hidden markov models
- real time
- linear prediction
- background noise
- pairwise
- visual attention model
- information retrieval