Unsupervised Training for Deep Speech Source Separation with Kullback-Leibler Divergence Based Probabilistic Loss Function.
Masahito TogamiYoshiki MasuyamaTatsuya KomatsuYu NakagomePublished in: ICASSP (2020)
Keyphrases
- loss function
- kullback leibler divergence
- information theoretic
- mutual information
- pairwise
- supervised learning
- support vector
- distance measure
- speech recognition
- unsupervised learning
- probability density function
- training set
- semi supervised
- generative model
- audio features
- audio visual
- similarity measure
- bayesian networks