Supervised Attention Multi-Scale Temporal Convolutional Network for monaural speech enhancement.
Zehua ZhangLu ZhangXuyi ZhuangYukun QianMingjiang WangPublished in: EURASIP J. Audio Speech Music. Process. (2024)
Keyphrases
- multiscale
- speech enhancement
- convolutional network
- coarse to fine
- convolutional neural networks
- noise reduction
- signal to noise ratio
- noisy environments
- speech signal
- edge detection
- single channel
- natural images
- linear prediction
- scale space
- information retrieval
- feature selection
- vocal tract
- wavelet transform
- image compression
- multiresolution
- wavelet decomposition
- computer vision
- image processing
- image segmentation
- lossless compression
- local binary pattern
- high resolution
- image registration