Using information theoretic distance measures for solving the permutation problem of blind source separation of speech signals.
Eugen HoffmannDorothea KolossaBert-Uwe KöhlerReinhold OrglmeisterPublished in: EURASIP J. Audio Speech Music. Process. (2012)
Keyphrases
- information theoretic
- blind source separation
- distance measure
- speech signal
- kullback leibler divergence
- mutual information
- information theory
- similarity measure
- independent component analysis
- non stationary
- speech recognition
- euclidean distance
- blind separation
- kullback leibler
- hidden markov models
- kl divergence
- negative matrix factorization
- pairwise
- multiscale
- sound source
- image processing
- neural network