Phonotactic Reconstruction of Encrypted VoIP Conversations: Hookt on Fon-iks

摘要: In this work, we unveil new privacy threats against Voice-over-IP (VoIP) communications. Although prior work has shown that the interaction of variable bit-rate codecs and length-preserving stream ciphers leaks information, show threat is more serious than previously thought. particular, derive approximate transcripts encrypted VoIP conversations by segmenting an observed packet into subsequences representing individual phonemes classifying those they encode. Drawing on insights from computational linguistics speech recognition communities, apply novel techniques for unmasking parts conversation. We believe our ability to do so underscores importance designing secure (yet efficient) ways protect confidentiality conversations.