mlsanityData to 7 Oct 2026

Membership Inference Attacks Against Self-supervised Speech Models

arXiv 2111.05113 ↗First posted 9 Nov 2021cs.CR cs.LG cs.SDcited by 16

On the map: LLM safety and security → Membership inference attacks

Abstract (arXiv)

Recently, adapting the idea of self-supervised learning (SSL) on continuous speech has started gaining attention. SSL models pre-trained on a huge amount of unlabeled audio can generate general-purpose representations that benefit a wide variety of speech processing tasks. Despite their ubiquitous deployment, however, the potential privacy risks of these models have not been well investigated. In this paper, we present the first privacy analysis on several SSL speech models using Membership Inference Attacks (MIA) under black-box access. The experiment results show that these pre-trained models are vulnerable to MIA and prone to membership information leakage with high Area Under the Curve (AUC) in both utterance-level and speaker-level. Furthermore, we also conduct several ablation studies to understand the factors that contribute to the success of MIA.

Key facts. Cited by 16 papers in the map (data through 7 Oct 2026). Placed in Membership inference attacks, one of 12 directions in LLM safety and security. Builds on 22 of its 35 references that are in the map. First posted on arXiv 9 Nov 2021.

Builds on

22 of 35 references.

References · 22

  1. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  2. 2020wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representationscited by 3,853
  3. 2021HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Unitscited by 2,301
  4. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,221
  5. 2020Extracting Training Data from Large Language Modelscited by 2,144
  6. 2021SUPERB: Speech processing Universal PERformance Benchmarkcited by 652
  7. 2016Stealing Machine Learning Models via Prediction APIscited by 602
  8. 2018Exploiting Unintended Feature Leakage in Collaborative Learningcited by 509
  9. 2018ML-Leaks: Model and Data Independent Membership Inference Attacks and Defenses on Machine Learning Modelscited by 475
  10. 2019Libri-Light: A Benchmark for ASR with Limited or No Supervisioncited by 462
  11. 2020Label-Only Membership Inference Attackscited by 371
  12. 2020Information Leakage in Embedding Modelscited by 199
10 more
  1. 2019Mockingjay: Unsupervised Speech Representation Learning with Deep Bidirectional Transformer Encoderscited by 196
  2. 2020TERA: Self-Supervised Learning of Transformer Encoder Representation for Speechcited by 186
  3. 2019An Unsupervised Autoregressive Model for Speech Representation Learningcited by 185
  4. 2020Multi-task self-supervised learning for Robust Speech Recognitioncited by 127
  5. 2020Unsupervised pretraining transfers well across languagescited by 100
  6. 2019Learning Problem-agnostic Speech Representations from Multiple Self-supervised Taskscited by 87
  7. 2021Node-Level Membership Inference Attacks Against Graph Neural Networkscited by 80
  8. 2020An Overview of Privacy in Machine Learningcited by 39
  9. 2021Quantifying and Mitigating Privacy Risks of Contrastive Learningcited by 35
  10. 2020Self-attention encoding and pooling for speaker recognitioncited by 33

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 2; 2023: 2; 2024: 5; 2025: 4; 2026: 3.

16 papers.

Sort:

By area:

  1. 2024Is My Data in Your Retrieval Database? Membership Inference Attacks Against Retrieval Augmented Generationcited by 44
  2. 2023SLMIA-SR: Speaker-Level Membership Inference Attacks against Speaker Recognition Systemscited by 14
  3. 2022A Review of Speech-centric Trustworthy Machine Learning: Privacy, Safety, and Fairnesscited by 13
  4. 2024Gibberish is All You Need for Membership Inference Detection in Contrastive Language-Audio Pretrainingcited by 10
  5. 2023Generation or Replication: Auscultating Audio Latent Diffusion Modelscited by 9
  6. 2024A Unified Membership Inference Method for Visual Self-supervised Encoder via Part-aware Capabilitycited by 9
  7. 2024Safety and Performance, Why Not Both? Bi-Objective Optimized Model Compression against Heterogeneous Attacks Toward AI Software Deploymentcited by 4
  8. 2024Improving Membership Inference in ASR Model Auditing with Perturbed Loss Featurescited by 2
  9. 2025Protecting Bystander Privacy via Selective Hearing in Audio LLMscited by 1
  10. 2025Assessing the Effectiveness of Membership Inference on Generative Musiccited by 1
  11. 2026Membership Inference Attacks against Large Audio Language Modelscited by 1
  12. 2022An Ensemble Teacher-Student Learning Approach with Poisson Sub-sampling to Differential Privacy Preserving Speech Recognitioncited by 0
4 more
  1. 2025Adversarial Attacks in Multimodal Systems: A Practitioner's Surveycited by 0
  2. 2025Personal Attribute Leakage in Federated Speech Modelscited by 0
  3. 2026Hearing the Whispers: Black-Box Membership Inference Attacks on Finetuned TTS Modelscited by 0
  4. 2026Reusing Latent Speech Representations for Query-Conditioned Topic Localization in Transcriptscited by 0