Group nonnegative matrix factorisation with speaker and session variability compensation for speaker identification.
Romain SerizelSlim EssidGaël RichardPublished in: ICASSP (2016)
Keyphrases
- speaker identification
- matrix factorisation
- gaussian mixture model
- em algorithm
- speaker recognition
- speech recognition
- speaker dependent
- noisy environments
- speech signal
- feature extraction
- feature transformation
- broadcast news
- speaker independent
- expectation maximization
- pairwise
- speaker diarization
- nonnegative matrix factorization
- automatic speech recognition
- maximum likelihood
- speaker verification
- image processing
- mixture model