Adaptation of Speaker-Specific Bases in Non-Negative Matrix Factorization for Single Channel Speech-Music Separation.
Emad M. GraisHakan ErdoganPublished in: INTERSPEECH (2011)
Keyphrases
- negative matrix factorization
- single channel
- speech enhancement
- speech recognition
- multi channel
- matrix factorization
- automatic speech recognition
- audio visual
- vocal tract
- document clustering
- sparse representation
- speech signal
- principal component analysis
- independent component analysis
- pattern recognition
- blind source separation
- audio features
- data mining
- data analysis
- noise reduction
- noisy environments
- information retrieval
- frequency domain
- machine learning