Leveraging Pseudo-labeled Data to Improve Direct Speech-to-Speech Translation.
Qianqian DongFengpeng YueTom KoMingxuan WangQibing BaiYu ZhangPublished in: CoRR (2022)
Keyphrases
- labeled data
- unlabeled data
- semi supervised
- semi supervised learning
- training data
- speech recognition
- active learning
- semi supervised classification
- text classification
- supervised learning
- domain adaptation
- data points
- prior knowledge
- learning algorithm
- transfer learning
- training examples
- labeled training data
- label propagation
- labeling effort
- text categorization
- class labels
- data sets
- co training
- labeled data for training
- labeled and unlabeled data
- collaborative filtering
- neural network