Disentangled Speech Representation Learning Based on Factorized Hierarchical Variational Autoencoder with Self-Supervised Objective.
Yuying XieThomas ArildsenZheng-Hua TanPublished in: MLSP (2021)
Keyphrases
- learning tasks
- reinforcement learning
- learning process
- knowledge acquisition
- learning algorithm
- hierarchical reinforcement learning
- language acquisition
- learning problems
- speech recognition
- supervised learning
- active learning
- background knowledge
- hierarchical structure
- learning systems
- online learning
- inductive inference
- dynamic bayesian networks
- visual representation