LITEVSR: Efficient Visual Speech Recognition by Learning from Speech Representations of Unlabeled Data.
Hendrik LauxEmil MededovicAhmed HallawaLukas MartinArne PeineAnke SchmeinkPublished in: ICASSP (2024)
Keyphrases
- unlabeled data
- learning algorithm
- active learning
- semi supervised learning
- semi supervised
- labeled and unlabeled data
- supervised learning
- labeled data
- co training
- visual speech recognition
- learning process
- domain adaptation
- prior knowledge
- positive examples
- unsupervised learning
- reinforcement learning
- class labels
- text classification
- training set
- training data
- machine learning
- data points
- training examples
- speech recognition
- computer vision