Cross-Domain Deep Visual Feature Generation for Mandarin Audio-Visual Speech Recognition.
Rongfeng SuXunying LiuLan WangJingzhou YangPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2020)
Keyphrases
- cross domain
- feature generation
- text categorization
- audio visual speech recognition
- transfer learning
- text classification
- knowledge transfer
- inductive learning
- speech recognition
- visual speech
- visual features
- feature selection
- knn
- audio visual
- visual information
- information extraction
- machine learning
- active learning
- document classification
- semi supervised learning
- k nearest neighbor
- information retrieval
- low level