Exploring the influence of fine-tuning data on wav2vec 2.0 model for blind speech quality prediction.
Helard Becerra MartinezAlessandro RaganoAndrew HinesPublished in: INTERSPEECH (2022)
Keyphrases
- quality prediction
- fine tuning
- experimental data
- data sets
- empirical data
- computational model
- data structure
- prior knowledge
- sensor data
- database
- test data
- mathematical model
- input data
- probability distribution
- image data
- data quality
- statistical methods
- data analysis
- synthetic data
- high level
- simulation data
- data collection
- data processing
- probabilistic model
- measured data
- learning models
- statistical model
- network structure
- expert knowledge
- data points
- original data
- raw data
- conceptual model
- data sources
- knowledge discovery
- high dimensional data
- computer systems
- expectation maximization
- maximum likelihood