Can DeepFake Speech be Reliably Detected?
Not on the map. Why?
Abstract (arXiv)
Recent advances in text-to-speech (TTS) systems, particularly those with voice cloning capabilities, have made voice impersonation readily accessible, raising ethical and legal concerns due to potential misuse for malicious activities like misinformation campaigns and fraud. While synthetic speech detectors (SSDs) exist to combat this, they are vulnerable to ``test domain shift", exhibiting decreased performance when audio is altered through transcoding, playback, or background noise. This vulnerability is further exacerbated by deliberate manipulation of synthetic speech aimed at deceiving detectors. This work presents the first systematic study of such active malicious attacks against state-of-the-art open-source SSDs. White-box attacks, black-box attacks, and their transferability are studied from both attack effectiveness and stealthiness, using both hardcoded metrics and human ratings. The results highlight the urgent need for more robust detection methods in the face of evolving adversarial threats.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 28 of its 42 references that are in the map. First posted on arXiv 9 Oct 2024.
Builds on
28 of 42 references.
References · 28
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
- 2016Adversarial examples in the physical worldcited by 1,769
- 2023Neural Codec Language Models are Zero-Shot Text to Speech Synthesizerscited by 752
- 2020FastSpeech 2: Fast and High-Quality End-to-End Text to Speechcited by 717
- 2022AudioLM: a Language Modeling Approach to Audio Generationcited by 616
- 2017Tacotron: Towards End-to-End Speech Synthesiscited by 490
- 2023Voicebox: Text-Guided Multilingual Universal Speech Generation at Scalecited by 379
- 2019MelGAN: Generative Adversarial Networks for Conditional Waveform Synthesiscited by 354
- 2022AudioGen: Textually Guided Audio Generationcited by 295
- 2021YourTTS: Towards Zero-Shot Multi-Speaker TTS and Zero-Shot Voice Conversion for everyonecited by 278
- 2019Simple Black-box Adversarial Attackscited by 251
- 2023NaturalSpeech 2: Latent Diffusion Models are Natural and Zero-Shot Speech and Singing Synthesizerscited by 235
16 more
- 2021AASIST: Audio Anti-Spoofing using Integrated Spectro-Temporal Graph Attention Networkscited by 180
- 2019ASVspoof 2019: Future Horizons in Spoofed and Fake Audio Detectioncited by 178
- 2020End-to-end anti-spoofing with RawNet2cited by 164
- 2022Does Audio Deepfake Detection Generalize?cited by 135
- 2024VALL-E 2: Neural Codec Language Models are Human Parity Zero-Shot Text to Speech Synthesizerscited by 122
- 2021WaveFake: A Data Set to Facilitate Audio Deepfake Detectioncited by 78
- 2019Zero-Shot Multi-Speaker Text-To-Speech with State-of-the-art Neural Speaker Embeddingscited by 73
- 2021End-to-End Spectro-Temporal Graph Attention Networks for Speaker Verification Anti-Spoofing and Speech Deepfake Detectioncited by 70
- 2023PromptTTS 2: Describing and Generating Voices with Text Promptcited by 48
- 2024The Codecfake Dataset and Countermeasures for the Universally Detection of Deepfake Audiocited by 31
- 2024FlashSpeech: Efficient Zero-Shot Speech Synthesiscited by 24
- 2024AudioMarkBench: Benchmarking Robustness of Audio Watermarkingcited by 23
- 2021Graph Attention Networks for Anti-Spoofingcited by 17
- 2022Virtuoso: Massive Multilingual Speech-Text Joint Semi-Supervised Learning for Text-To-Speechcited by 15
- 2024Meta Learning Text-to-Speech Synthesis in over 7000 Languagescited by 9
- 2024Zero-shot Cross-lingual Voice Transfer for TTScited by 1
Built on by
Citing papers per year: 2025: 1.
1 paper.