Exploring Audio Editing Features as User-Centric Privacy Defenses Against Large Language Model(LLM) Based Emotion Inference Attacks
This addresses privacy concerns for users of speech-enabled technologies, offering a practical defense against emotion inference attacks, though it is incremental as it builds on existing audio editing techniques.
The paper tackles the problem of protecting emotional privacy in audio data from inference attacks by LLMs and other models, introducing a user-centric approach using pitch and tempo manipulation that effectively obfuscates emotional data across three datasets.
The rapid proliferation of speech-enabled technologies, including virtual assistants, video conferencing platforms, and wearable devices, has raised significant privacy concerns, particularly regarding the inference of sensitive emotional information from audio data. Existing privacy-preserving methods often compromise usability and security, limiting their adoption in practical scenarios. This paper introduces a novel, user-centric approach that leverages familiar audio editing techniques, specifically pitch and tempo manipulation, to protect emotional privacy without sacrificing usability. By analyzing popular audio editing applications on Android and iOS platforms, we identified these features as both widely available and usable. We rigorously evaluated their effectiveness against a threat model, considering adversarial attacks from diverse sources, including Deep Neural Networks (DNNs), Large Language Models (LLMs), and and reversibility testing. Our experiments, conducted on three distinct datasets, demonstrate that pitch and tempo manipulation effectively obfuscates emotional data. Additionally, we explore the design principles for lightweight, on-device implementation to ensure broad applicability across various devices and platforms.