Supervised Audio Source Separation Based on Nonnegative Matrix Factorization with Cosine Similarity Penalty.
Yuta IwaseDaichi KitamuraPublished in: IEICE Trans. Fundam. Electron. Commun. Comput. Sci. (2022)
Keyphrases
- nonnegative matrix factorization
- audio features
- document clustering
- negative matrix factorization
- matrix factorization
- multimedia
- semi supervised
- data representation
- audio visual
- low level
- supervised learning
- principal component analysis
- spectral clustering
- clustering algorithm
- semantic similarity
- distance measure
- data mining
- machine learning
- document collections
- original data
- signal processing
- feature set
- text mining
- clustering method
- unsupervised learning
- multiscale
- feature selection