High-Frequency Adversarial Defense for Speech and Audio
2021
Recent work suggests that adversarial examples are enabled by high-frequency components in the dataset. In the speech domain where spectrograms are used extensively, masking those components seems like a sound direction for defenses against attacks. We explore a smoothing approach based on additive noise masking in priority high frequencies. We show that this approach is much more robust than the naive noise filtering approach, and a promising research direction. We successfully apply our defense on a Librispeech speaker identification task, and on the UrbanSound8K audio classification dataset.
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
23
References
1
Citations
NaN
KQI