Continuous Metric Learning For Transferable Speech Emotion Recognition and Embedding Across Low-resource Languages.
Sneha DasNicklas Leander LundNicole Nadine LønfeldtAnne Katrine PagsbergLine H. ClemmensenPublished in: CoRR (2022)
Keyphrases
- metric learning
- speech emotion recognition
- maximum variance unfolding
- distance metric learning
- distance metric
- multi task
- learning tasks
- semi supervised
- machine learning and pattern recognition
- dimensionality reduction
- pairwise
- distance function
- kernel matrix
- nonlinear dimensionality reduction
- semi supervised clustering
- feature space
- data sets
- mahalanobis metric
- subject to linear constraints
- laplacian eigenmaps
- graph embedding
- manifold learning
- high dimensional data
- multi class
- domain knowledge
- similarity measure