What Do Self-Supervised Speech and Speaker Models Learn? New Findings from a Cross Model Layer-Wise Analysis.
Takanori AshiharaMarc DelcroixTakafumi MoriyaKohei MatsuuraTaichi AsamiYusuke IjimaPublished in: ICASSP (2024)
Keyphrases
- probabilistic model
- empirical data
- autoregressive
- statistical model
- computational model
- experimental data
- statistical models
- structural models
- modeling framework
- computational models
- theoretical models
- generic model
- linear model
- classification models
- parametric models
- prior knowledge
- speech signal
- learned models
- structural model
- model validation
- mathematical model
- speech recognition
- probability distribution
- high level
- audio visual
- prediction model
- neural network model
- analytical model
- process model
- parameter estimation
- multiple models
- structural equation modeling
- model selection
- efficient inference and learning