Introducing Model Inversion Attacks on Automatic Speaker Recognition

Abstract

Model inversion (MI) attacks allow to reconstruct average per-class representations of a machine learning (ML) model's training data. It has been shown that in scenarios where each class corresponds to a different individual, such as face classifiers, this represents a severe privacy risk. In this work, we explore a new application for MI: the extraction of speakers' voices from a speaker recognition system. We present an approach to (1) reconstruct audio samples from a trained ML model and (2) extract intermediate voice feature representations which provide valuable insights into the speakers' biometrics. Therefore, we propose an extension of MI attacks which we call sliding model inversion. Our sliding MI extends standard MI by iteratively inverting overlapping chunks of the audio samples and thereby leveraging the sequential properties of audio data for enhanced inversion performance. We show that one can use the inverted audio data to generate spoofed audio samples to impersonate a speaker, and execute voice-protected commands for highly secured systems on their behalf. To the best of our knowledge, our work is the first one extending MI attacks to audio data, and our results highlight the security risks resulting from the extraction of the biometric data in that setup.

arXiv commentfor associated pdf, see https://www.isca-speech.org/archive/pdfs/spsc_2022/pizzi22_spsc.pdf

References in corpus (7)

• WaveNet: A Generative Model for Raw Audio
• ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification
• Transfer Learning from Speaker Verification to Multispeaker Text-To-Speech Synthesis
• Updates-Leak: Data Set Inference and Reconstruction Attacks in Online Learning
• Evaluating Voice Conversion-based Privacy Protection against Informed Attackers
• Defending Model Inversion and Membership Inference Attacks via Prediction Purification
• A Tandem Framework Balancing Privacy and Security for Voice User Interfaces

Cited by in corpus (2)

• SLMIA-SR: Speaker-Level Membership Inference Attacks against Speaker Recognition Systems
• Adversarial Attacks in Multimodal Systems: A Practitioner's Survey