Speech enhancement using segmental nonnegative matrix factorization.
Hao-Teng FanJeih-weih HungXugang LuSyu-Siang WangYu TsaoPublished in: ICASSP (2014)
Keyphrases
- nonnegative matrix factorization
- speech enhancement
- noise reduction
- noisy environments
- signal to noise ratio
- single channel
- speech signal
- hidden markov models
- data representation
- negative matrix factorization
- matrix factorization
- least squares
- linear prediction
- multi channel
- vocal tract
- smoothing algorithm
- speech recognition
- principal component analysis
- wiener filter
- document clustering
- additive noise
- spectral clustering
- automatic speech recognition
- image acquisition
- image coding
- image processing
- objective function
- data sets
- original data
- sparse representation
- edge detection
- collaborative filtering
- semi supervised