Multi-layer encoder-decoder time-domain single channel speech separation.
Debang LiuTianqi ZhangMads Græsbøll ChristensenChen YiYing WeiPublished in: Pattern Recognit. Lett. (2024)
Keyphrases
- multi layer
- single channel
- speech enhancement
- frequency domain
- video codec
- distributed video coding
- multi channel
- sound source
- speech signal
- single layer
- rate distortion
- video coding
- motion estimation
- bit rate
- neural network
- spatial domain
- speech recognition
- audio visual
- neural nets
- prior information
- independent component analysis
- subband
- noisy environments
- signal processing
- noise reduction
- brain activity
- wavelet decomposition
- non stationary
- visual information
- hidden markov models
- feature vectors
- feature extraction