Deep Recurrent NMF for Speech Separation by Unfolding Iterative Thresholding.
Scott WisdomThomas PowersJames W. PittonLes AtlasPublished in: CoRR (2017)
Keyphrases
- negative matrix factorization
- nonnegative matrix factorization
- speech recognition
- edge detection
- matrix factorization
- image processing
- spoken language
- post processing
- image segmentation
- speech signal
- audio visual
- petri net
- feed forward
- gray level
- principal component analysis
- dialogue system
- speech synthesis
- endpoint detection
- hearing impaired
- multi unit combinatorial auctions
- recurrent neural networks
- automatic speech recognition
- deep learning
- objective function
- text to speech
- recognition engine