Overdetermined blind separation for convolutive mixtures of speech based on multistage ICA using subarray processing.
Tsuyoki NishikawaHiroshi AbeHiroshi SaruwatariKiyohiro ShikanoPublished in: ICASSP (1) (2004)
Keyphrases
- multistage
- blind separation
- blind source separation
- convolutive mixtures
- speech signal
- independent component analysis
- non stationary
- dynamic programming
- source separation
- speech recognition
- source signals
- independent components
- optimal policy
- factor analysis
- audio visual
- negative matrix factorization
- sparse representation
- signal processing
- pattern recognition