Error Modeling via Asymmetric Laplace Distribution for Deep Neural Network Based Single-Channel Speech Enhancement.
Li ChaiJun DuChin-Hui LeePublished in: INTERSPEECH (2018)
Keyphrases
- single channel
- speech enhancement
- multi channel
- noise reduction
- signal to noise ratio
- noisy environments
- independent component analysis
- frequency domain
- wiener filter
- wavelet decomposition
- speech signal
- prior information
- signal processing
- smoothing algorithm
- vocal tract
- linear prediction
- brain activity
- feature extraction
- sound source
- random variables
- multiresolution