Blind extraction of a dominant source from mixtures of many sources using ICA and time-frequency masking.
Hiroshi SawadaShoko ArakiRyo MukaiShoji MakinoPublished in: ISCAS (6) (2005)
Keyphrases
- source signals
- independent component analysis
- blind source separation
- source separation
- blind separation
- signal processing
- independent components
- principal component analysis
- blind deconvolution
- single channel
- factor analysis
- nonlinear functions
- frequency domain
- non stationary
- information extraction
- motion deblurring
- multiple sources
- automatic extraction
- temporal structure
- feature extraction
- databases
- independent components analysis
- single source
- filter responses
- negative matrix factorization
- automatically extracted
- information loss
- human visual system
- fourier transform
- website
- neural network