Combining non-negative matrix factorization and deep neural networks for speech enhancement and automatic speech recognition.
Thanh T. VuBenjamin BigotEng Siong ChngPublished in: ICASSP (2016)
Keyphrases
- negative matrix factorization
- automatic speech recognition
- speech signal
- speech enhancement
- noisy environments
- neural network
- speech recognition
- vocal tract
- pattern recognition
- matrix factorization
- document clustering
- background noise
- sparse representation
- principal component analysis
- linear prediction
- broadcast news
- hidden markov models
- noise reduction
- non stationary
- acoustic features
- spectral clustering
- sound source
- language model
- speech synthesis
- signal to noise ratio
- collaborative filtering
- multiresolution
- image processing