1 paper
Evgeny Kushnir, Alexandr Kozodaev, Dmitrii Korzh +3
Recent advances in generative models have amplified the risk of malicious misuse of speech synthesis technologies, enabling adversaries to impersonate target speakers and access se…