Localization based stereo speech source separation using probabilistic time-frequency masking and deep neural networks.
Yang YuWenwu WangPeng HanPublished in: EURASIP J. Audio Speech Music. Process. (2016)
Keyphrases
- source separation
- neural network
- audio features
- independent component analysis
- blind source separation
- pattern recognition
- speech recognition
- computer vision
- denoising
- audio visual
- speech signal
- probabilistic model
- wavelet transform
- temporal structure
- single channel
- generative model
- human visual system
- spatio temporal
- factor analysis
- multi modal
- wavelet decomposition
- high level