Boosting Noise Robustness of Acoustic Model via Deep Adversarial Training.
Bin LiuShuai NieYaping ZhangDengfeng KeShan LiangWenju LiuPublished in: ICASSP (2018)
Keyphrases
- base classifiers
- training set
- training samples
- training error
- ensemble learning
- greater robustness
- supervised learning
- random noise
- test set
- noise level
- training and testing data
- noise immunity
- image noise
- training data
- learning algorithm
- training examples
- training phase
- training process
- online learning
- additive noise
- gaussian noise
- noise reduction
- early stopping
- salt pepper
- generalization error
- deep architectures
- noise model
- combining multiple
- noisy data
- multi agent
- image processing