Think your speech model is secure? It might be quietly leaking what it was trained on. In a new blog post, we explain membership inference attacks and why they matter for cyber security experts. 🔗 neodyme.io/en/blog/memb...
Did You Train on My Voice? Exploring Privacy Risks in ASR
This post explores a recent research paper on membership inference attacks targeting Automatic Speech Recognition (ASR) models. It breaks down how subtle signals like input perturbation and model loss...
neodyme.io