An EM algorithm for joint source separation and diarisation of multichannel convolutive speech mixtures.
Dionyssos Kounades-BastianLaurent GirinXavier Alameda-PinedaSharon GannotRadu HoraudPublished in: ICASSP (2017)
Keyphrases
- source separation
- em algorithm
- single channel
- expectation maximization
- multi channel
- audio features
- blind source separation
- mixture model
- independent component analysis
- maximum likelihood
- speech signal
- speech recognition
- generative model
- gaussian mixture model
- gaussian mixture
- denoising
- source signals
- audio visual
- hyperparameters
- incomplete data
- frequency domain
- gaussian process
- mixture distribution
- signal processing
- principal component analysis
- prior information
- multi modal
- feature set
- low dimensional
- sound source
- text mining
- wiener filter
- temporal structure
- low level
- prior knowledge
- multiresolution