Skip to content

Hearing the Whispers: Black-Box Membership Inference Attacks on Finetuned TTS Models

Sep 2026 · 0 citations · 78 references
Computer Science Engineering

TL;DR

This work presents the first black-box MIA framework explicitly tailored to TTS models at both the speaker and record levels, and characterize the feasible query space and establish two criteria, scorable extent and memorization elicitation, for evaluating five representative queries.

Abstract

Text-to-Speech (TTS) foundation models are increasingly fine-tuned on private datasets to synthesize highly personalized voices, introducing severe privacy risks by exposing both biometric identities and sensitive speech content. Existing black-box membership inference attacks (MIAs) follow a two-stage pipeline of query generation and representation engineering, both of which face unique challenges when adapted to TTS. For query generation, dual conditioning on synthesis text and reference speech creates a large and underexplored query design space with no established criterion for identifying an effective query. For representation engineering, the multi-level speech characteristics and temporal variability of speech make low-level representations and direct comparisons inadequate for capturing membership signals. To address these challenges, we present the first black-box MIA framework explicitly tailored to TTS models at both the speaker and record levels. For query generation, we characterize the feasible query space and establish two criteria, scorable extent and memorization elicitation, for evaluating five representative queries, identifying recitation as the strongest. For representation engineering, we obtain multi-level speech representations from embedding models and temporally align the generated and target audio for fine-grained comparison. Evaluations across three state-of-the-art TTS models (CosyVoice2, F5-TTS, and XTTS-v2) fine-tuned on two benchmark datasets (VCTK and British Dialect) reveal severe privacy leakage: speaker-level AUC remains above 0.80 and approaches 1.0 in the strongest settings, while record-level AUC ranges from 0.80 to 0.90 and remains effective even in challenging scenarios where both members and non-members are of the same speakers. We further identify speech characteristics associated with disproportionate vulnerability to memorization.

View source

Similar papers

Open access Sep 2026

TTS-Guard: Black-Box Ownership Verification of Text-to-Speech Models via Adaptive Adversarial Speaker-Pair Fingerprints

TTS-Guard is presented, a black-box ownership verification framework for TTS models built on adversarial speaker-pair fingerprints that selects key speaker pairs in a dual embedding space for architecture-agnostic stealth and optimises a perturbation through an adaptive curriculum of shadow models covering fine-tuning,...

Xu-Bin Yue, Zhenhua Xu, Zhebo Wang et al. · 0 citations
Preprint Sep 2026

Silence-the-Mimic: Accelerating Imperceptible Perturbation Generation Against Voice Cloning

Deep neural network-based Voice Conversion (VC) and Text-to-Speech (TTS) models have rapidly advanced, enabling realistic voice cloning with minimal input data. Such capabilities raise serious concerns over unauthorized cloning of speaker identities and the associated privacy and security risks. Current imperceptible a...

Run-Qiu Xu · 0 citations
Sep 2026

GenTraceBench: A Benchmark for Tracing Audio Deepfakes Across Pre- and Post-training Stages

Modern text-to-speech (TTS) systems are rarely deployed as unchanged pre-trained models. They are often adapted through supervised fine-tuning (SFT) or preference optimization such as DPO and GRPO. This raises a practical question for audio deepfake forensics: do fingerprints learned from a foundation generator remain...

Li Wang, Kun-Yu Feng, Wan Lin et al. · 0 citations
#artificial intelligence Preprint Sep 2026

Fingerprinting Multimodal Large Language Models

While multimodal large language models (MLLMs) enable a wide range of image-text reasoning tasks, recent incidents indicate that they are vulnerable to illicit deployment and unauthorized distillation. Existing solutions for model provenance are typically confounded by shared language backbones in MLLMs and struggle to...

Chao Huang, Meng Tong, Kejiang Chen · 0 citations
Preprint Sep 2026

GhostWord: A Fine-Grained Backdoor Attack on Automatic Speech Recognition

During poisoning, a trigger is injected into the forced-aligned time span of a chosen source word in the audio and replace only that word in the transcript, enabling precise semantic flips and composable sentence manipulation while avoiding many-to-one label artifacts.

Mojtaba Nafez, Mobina Poulaei, Kiarash Kiani Feriz et al. · 0 citations

Related blog posts

Microsoft Research Blog Nov 3, 2026

Learn the rules: The hacker mindset

Research Manager Mike Walker reflects on a career spanning hacker culture, cybersecurity, and AI, sharing lessons about trust, teamwork, leadership, and why solving difficult problems requires curiosity, judgment, agility—and an understanding of the rules. The post Learn the rules: The hacker mindset appeared first on Microsoft Research.

MIT News · Artificial Intelligence Nov 17, 2026

Learning the language of machine learning

By studying the computational foundations of language learning, Associate Professor Jacob Andreas builds intelligent systems that can communicate effectively with humans.

Microsoft Research Blog Nov 16, 2026

TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient

AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.