Deep versus Wide: An Analysis of Student Architectures for Task-Agnostic Knowledge Distillation of Self-Supervised Speech Models.
Takanori AshiharaTakafumi MoriyaKohei MatsuuraTomohiro TanakaPublished in: CoRR (2022)
Keyphrases
- prior knowledge
- empirical data
- statistical models
- domain knowledge
- speech recognition
- knowledge level
- expert systems
- learning process
- machine learning
- learning styles
- language model
- student learning
- computational models
- complex systems
- domain experts
- statistical analysis
- knowledge management
- knowledge discovery
- knowledge representation
- data analysis
- learning environment
- information retrieval