Supervised non-euclidean sparse NMF via bilevel optimization with applications to speech enhancement.
Pablo SprechmannAlexander M. BronsteinGuillermo SapiroPublished in: HSCMA (2014)
Keyphrases
- speech enhancement
- negative matrix factorization
- noisy environments
- signal to noise ratio
- noise reduction
- single channel
- nonnegative matrix factorization
- feature selection
- least squares
- speech signal
- sparse representation
- frequency domain
- probabilistic model
- linear prediction
- multiresolution
- vocal tract
- smoothing algorithm
- image sequences