Self-Supervised Audio-and-Text Pre-training with Extremely Low-Resource Parallel Data.
Yu KangTianqiao LiuHang LiYang HaoWenbiao DingPublished in: CoRR (2022)
Keyphrases
- data sets
- data collection
- data sources
- learning algorithm
- high quality
- data analysis
- data processing
- data mining techniques
- training examples
- database
- probability distribution
- training data
- input data
- synthetic data
- original data
- text data
- textual data
- image data
- end users
- training set
- multimedia
- multimedia data
- raw data
- data quality