Blind extraction of a dominant source signal from mixtures of many sources [audio source separation applications].
Hiroshi SawadaShoko ArakiRyo MukaiShoji MakinoPublished in: ICASSP (3) (2005)
Keyphrases
- source separation
- source signals
- audio features
- blind source separation
- independent component analysis
- single channel
- audio visual
- signal processing
- blind separation
- feature set
- non stationary
- denoising
- information extraction
- visual features
- music retrieval
- multi channel
- multimedia
- low level
- temporal structure
- audio signal
- music information retrieval
- gaussian process
- sound source
- sparse representation
- pattern recognition
- high level