Adversarial Privacy Protection on Speech Enhancement. (arXiv:2206.08170v1 [cs.SD])

Web: http://arxiv.org/abs/2206.08170

Speech is easily leaked imperceptibly, such as being recorded by mobile
phones in different situations. Private content in speech may be maliciously
extracted through speech enhancement technology. Speech enhancement technology
has developed rapidly along with deep neural networks (DNNs), but adversarial
examples can cause DNNs to fail. In this work, we propose an adversarial method
to degrade speech enhancement systems. Experimental results show that generated
adversarial examples can erase most content information in original examples or
replace it with target …

arxiv on privacy speech