Unsupervised Robust Speech Enhancement Based on Alpha-Stable Fast Multichannel Nonnegative Matrix Factorization.
Mathieu FontaineKouhei SekiguchiAditya Arie NugrahaKazuyoshi YoshiiPublished in: INTERSPEECH (2020)
Keyphrases
- nonnegative matrix factorization
- speech enhancement
- single channel
- noisy environments
- data representation
- matrix factorization
- linear prediction
- noise reduction
- negative matrix factorization
- unsupervised learning
- least squares
- data sets
- spectral clustering
- speech signal
- semi supervised
- document clustering
- maximum likelihood
- pairwise
- machine learning