Kernel Machines Beat Deep Neural Networks on Mask-Based Single-Channel Speech Enhancement.
Like HuiSiyuan MaMikhail BelkinPublished in: INTERSPEECH (2019)
Keyphrases
- speech enhancement
- single channel
- kernel machines
- neural network
- multi channel
- support vector
- wiener filter
- noise reduction
- prior information
- pattern recognition
- semi supervised
- noisy environments
- independent component analysis
- wavelet decomposition
- signal to noise ratio
- frequency domain
- brain activity
- speech signal
- speech recognition
- signal processing
- support vector machine
- feature vectors
- multiscale
- training data
- similarity measure
- vocal tract
- machine learning