An Improved Fully Convolutional Network Based on Post-Processing with Global Variance Equalization and Noise-Aware Training for Speech Enhancement.
Wenlong LiKaoru HirotaYaping DaiZhiyang JiaPublished in: J. Adv. Comput. Intell. Intell. Informatics (2021)
Keyphrases
- post processing
- speech enhancement
- noisy environments
- noise reduction
- signal to noise ratio
- single channel
- preprocessing
- filtering method
- background noise
- speech signal
- vocal tract
- wiener filter
- additive noise
- multi channel
- median filter
- linear prediction
- speech recognition
- smoothing algorithm
- post processed
- training set
- noise level
- edge detection
- multiscale
- computer vision