Single-channel mixed speech recognition using deep neural networks.
Chao WengDong YuMichael L. SeltzerJasha DroppoPublished in: ICASSP (2014)
Keyphrases
- single channel
- speech enhancement
- neural network
- multi channel
- independent component analysis
- pattern recognition
- wiener filter
- wavelet decomposition
- speech signal
- signal to noise ratio
- prior information
- noise reduction
- frequency domain
- sound source
- noisy environments
- speech recognition
- principal component analysis
- non stationary
- signal processing
- image segmentation
- image processing