Hear No Evil: Towards Adversarial Robustness of Automatic Speech Recognition via Multi-Task Learning.
Nilaksh DasDuen Horng ChauPublished in: CoRR (2022)
Keyphrases
- automatic speech recognition
- multi task learning
- multi task
- speech recognition
- gaussian processes
- speech signal
- multiple tasks
- broadcast news
- learning tasks
- hidden markov models
- multitask learning
- learning problems
- transfer learning
- conversational speech
- high order
- theoretical analysis
- learning algorithm
- learning models
- inductive learning
- metric learning
- multi class
- information extraction
- neural network