Discriminative and reconstructive basis training for audio source separation with semi-supervised nonnegative matrix factorization.
Daichi KitamuraNobutaka OnoHiroshi SaruwatariYu TakahashiKazunobu KondoPublished in: IWAENC (2016)
Keyphrases
- semi supervised
- nonnegative matrix factorization
- source separation
- supervised learning
- audio features
- negative matrix factorization
- semi supervised learning
- labeled data
- blind source separation
- unlabeled data
- matrix factorization
- pairwise
- data representation
- active learning
- unsupervised learning
- independent component analysis
- training set
- least squares
- spectral clustering
- audio visual
- generative model
- signal processing
- principal component analysis
- multimedia
- document clustering
- feature extraction
- data sets
- denoising