Investigating Active-Learning-Based Training Data Selection for Speech Spoofing Countermeasure.
Xin WangJunichi YamagishiPublished in: SLT (2022)
Keyphrases
- active learning
- training data
- sample selection
- training set
- annotation effort
- labeled data
- supervised learning
- unlabeled data
- learning algorithm
- training examples
- generalization error
- semi supervised
- data sets
- random selection
- countermeasures
- speech recognition
- test set
- machine learning
- decision trees
- classification accuracy
- support vector machine
- selection strategy
- random sampling
- speech signal
- training samples
- speech synthesis
- labeled examples
- audio visual
- training process
- classification models
- test data
- semi supervised learning
- learning strategies
- relevance feedback
- noisy data
- batch mode
- prior knowledge
- naive bayes
- labeled data for training