Subband based blind source separation for convolutive mixtures of speech.
Shoko ArakiShoji MakinoRobert AichnerTsuyoki NishikawaHiroshi SaruwatariPublished in: ICASSP (5) (2003)
Keyphrases
- subband
- convolutive mixtures
- frequency domain
- blind source separation
- speech signal
- filter bank
- wavelet transform
- bit rate
- image compression
- spatial domain
- multiresolution
- speech recognition
- high frequency
- wavelet coefficients
- feature vectors
- discrete wavelet transform
- low pass
- wavelet decomposition
- wavelet domain
- audio visual
- blind separation
- additive noise
- independent component analysis
- non stationary
- image classification
- image segmentation
- hidden markov models