# DeepSonar: Towards Effective and Robust Detection of AI-Synthesized Fake Voices

arXiv: https://arxiv.org/abs/2005.13770 · HTML: https://mlsanity.com/paper/2005.13770/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 May 2020.

## Built on by (most cited first, 24 of 24)

- [ADD 2022: the First Audio Deep Synthesis Detection Challenge](https://mlsanity.com/paper/2202.08433/) (2022)
- [Deepfakes Generation and Detection: State-of-the-art, open challenges, countermeasures, and way forward](https://mlsanity.com/paper/2103.00484/) (2021)
- [Neural Analysis and Synthesis: Reconstructing Speech from Self-Supervised Representations](https://mlsanity.com/paper/2110.14513/) (2021)
- [Countering Malicious DeepFakes: Survey, Battleground, and Horizon](https://mlsanity.com/paper/2103.00218/) (2021)
- [Multimodal Automated Fact-Checking: A Survey](https://mlsanity.com/paper/2305.13507/) (2023)
- [Half-Truth: A Partially Fake Audio Detection Dataset](https://mlsanity.com/paper/2104.03617/) (2021)
- [The Ethical Implications of Generative Audio Models: A Systematic Literature Review](https://mlsanity.com/paper/2307.05527/) (2023)
- ["Hello, It's Me": Deep Learning-based Speech Synthesis Attacks in the Real World](https://mlsanity.com/paper/2109.09598/) (2021)
- [Deepfake Media Generation and Detection in the Generative AI Era: A Survey and Outlook](https://mlsanity.com/paper/2411.19537/) (2024)
- [NVC-Net: End-to-End Adversarial Voice Conversion](https://mlsanity.com/paper/2106.00992/) (2021)
- [CLAD: Robust Audio Deepfake Detection Against Manipulation Attacks with Contrastive Learning](https://mlsanity.com/paper/2404.15854/) (2024)
- [SceneFake: An Initial Dataset and Benchmarks for Scene Fake Audio Detection](https://mlsanity.com/paper/2211.06073/) (2022)
- [VoiceWukong: Benchmarking Deepfake Voice Detection](https://mlsanity.com/paper/2409.06348/) (2024)
- [Catch You and I Can: Revealing Source Voiceprint Against Voice Conversion](https://mlsanity.com/paper/2302.12434/) (2023)
- [Passive Deepfake Detection Across Multi-modalities: A Comprehensive Survey](https://mlsanity.com/paper/2411.17911/) (2024)
- [Sound Check: Auditing Audio Datasets](https://mlsanity.com/paper/2410.13114/) (2024)
- [Automated Deepfake Detection](https://mlsanity.com/paper/2106.10705/) (2021)
- [Defend Data Poisoning Attacks on Voice Authentication](https://mlsanity.com/paper/2209.04547/) (2022)
- [MFAAN: Unveiling Audio Deepfakes with a Multi-Feature Authenticity Network](https://mlsanity.com/paper/2311.03509/) (2023)
- [DF-Captcha: A Deepfake Captcha for Preventing Fake Calls](https://mlsanity.com/paper/2208.08524/) (2022)
- [Synthetic Voice Detection and Audio Splicing Detection using SE-Res2Net-Conformer Architecture](https://mlsanity.com/paper/2210.03581/) (2022)
- [Ethical and Technical Limits of Deepfake Speech Datasets](https://mlsanity.com/paper/2606.10911/) (2026)
- [Deep Fake Detection, Deterrence and Response: Challenges and Opportunities](https://mlsanity.com/paper/2211.14667/) (2022)
- [Explainable-by-Design Audio Deepfake Detection via Wiener-Hopf Linear Prediction](https://mlsanity.com/paper/2607.12584/) (2026)

## Abstract (arXiv)

With the recent advances in voice synthesis, AI-synthesized fake voices are indistinguishable to human ears and widely are applied to produce realistic and natural DeepFakes, exhibiting real threats to our society. However, effective and robust detectors for synthesized fake voices are still in their infancy and are not ready to fully tackle this emerging threat. In this paper, we devise a novel approach, named DeepSonar, based on monitoring neuron behaviors of speaker recognition (SR) system, ie, a deep neural network (DNN), to discern AI-synthesized fake voices. Layer-wise neuron behaviors provide an important insight to meticulously catch the differences among inputs, which are widely employed for building safety, robust, and interpretable DNNs. In this work, we leverage the power of layer-wise neuron activation patterns with a conjecture that they can capture the subtle differences between real and AI-synthesized fake voices, in providing a cleaner signal to classifiers than raw inputs. Experiments are conducted on three datasets (including commercial products from Google, Baidu, etc) containing both English and Chinese languages to corroborate the high detection rates (98.1% average accuracy) and low false alarm rates (about 2% error rate) of DeepSonar in discerning fake voices. Furthermore, extensive experimental results also demonstrate its robustness against manipulation attacks (eg, voice conversion and additive real-world noises). Our work further poses a new insight into adopting neuron behaviors for effective and robust AI aided multimedia fakes forensics as an inside-out approach instead of being motivated and swayed by various artifacts introduced in synthesizing fakes.