Unsupervised single-channel speech separation via deep neural network for different gender mixtures.
Yannan WangJun DuLi-Rong DaiChin-Hui LeePublished in: APSIPA (2016)
Keyphrases
- single channel
- speech enhancement
- source separation
- neural network
- multi channel
- sound source
- speech signal
- independent component analysis
- audio features
- blind source separation
- audio visual
- unsupervised learning
- speech recognition
- pattern recognition
- wavelet decomposition
- frequency domain
- wiener filter
- vocal tract
- prior information
- noisy environments
- brain activity
- noise reduction
- signal processing
- multiresolution
- automatic speech recognition
- multi modal
- signal to noise ratio
- heart rate
- feature space
- speech synthesis