# Continual Learning for Fake Audio Detection

arXiv: https://arxiv.org/abs/2104.07286 · HTML: https://mlsanity.com/paper/2104.07286/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 31 references that are in the map.
- First posted on arXiv 15 Apr 2021.

## Builds on (most cited first, 13 of 13)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Learning without Forgetting](https://mlsanity.com/paper/1606.09282/) (2016)
- [iCaRL: Incremental Classifier and Representation Learning](https://mlsanity.com/paper/1611.07725/) (2016)
- [Progressive Neural Networks](https://mlsanity.com/paper/1606.04671/) (2016)
- [Continual Lifelong Learning with Neural Networks: A Review](https://mlsanity.com/paper/1802.07569/) (2018)
- [Continual Learning with Deep Generative Replay](https://mlsanity.com/paper/1705.08690/) (2017)
- [End-to-End Incremental Learning](https://mlsanity.com/paper/1807.09536/) (2018)
- [ASVspoof 2019: Future Horizons in Spoofed and Fake Audio Detection](https://mlsanity.com/paper/1904.05441/) (2019)
- [STC Antispoofing Systems for the ASVspoof2019 Challenge](https://mlsanity.com/paper/1904.05576/) (2019)
- [Residual Continual Learning](https://mlsanity.com/paper/2002.06774/) (2020)
- [Knowledge Distillation For Recurrent Neural Network Language Modeling With Trust Regularization](https://mlsanity.com/paper/1904.04163/) (2019)
- [Multi-Task Siamese Neural Network for Improving Replay Attack Detection](https://mlsanity.com/paper/2002.07629/) (2020)

## Built on by (most cited first, 27 of 27)

- [ADD 2022: the First Audio Deep Synthesis Detection Challenge](https://mlsanity.com/paper/2202.08433/) (2022)
- [Recent Advances of Multimodal Continual Learning: A Comprehensive Survey](https://mlsanity.com/paper/2410.05352/) (2024)
- [What to Remember: Self-Adaptive Continual Learning for Audio Deepfake Detection](https://mlsanity.com/paper/2312.09651/) (2023)
- [Partially Fake Audio Detection by Self-attention-based Fake Span Discovery](https://mlsanity.com/paper/2202.06684/) (2022)
- [An Initial Investigation for Detecting Vocoder Fingerprints of Fake Audio](https://mlsanity.com/paper/2208.09646/) (2022)
- [Do You Remember? Overcoming Catastrophic Forgetting for Fake Audio Detection](https://mlsanity.com/paper/2308.03300/) (2023)
- [Fully Automated End-to-End Fake Audio Detection](https://mlsanity.com/paper/2208.09618/) (2022)
- [Multi-Dataset Co-Training with Sharpness-Aware Optimization for Audio Anti-spoofing](https://mlsanity.com/paper/2305.19953/) (2023)
- [Learning Representations for New Sound Classes With Continual Self-Supervised Learning](https://mlsanity.com/paper/2205.07390/) (2022)
- [Prompt Tuning for Audio Deepfake Detection: Computationally Efficient Test-time Domain Adaptation with Limited Target Dataset](https://mlsanity.com/paper/2410.09869/) (2024)
- [Continual Audio-Visual Sound Separation](https://mlsanity.com/paper/2411.02860/) (2024)
- [SceneFake: An Initial Dataset and Benchmarks for Scene Fake Audio Detection](https://mlsanity.com/paper/2211.06073/) (2022)
- [Adaptive Fake Audio Detection with Low-Rank Model Squeezing](https://mlsanity.com/paper/2306.04956/) (2023)
- [Region-Based Optimization in Continual Learning for Audio Deepfake Detection](https://mlsanity.com/paper/2412.11551/) (2024)
- [Catch You and I Can: Revealing Source Voiceprint Against Voice Conversion](https://mlsanity.com/paper/2302.12434/) (2023)
- [Generalizable speech deepfake detection via meta-learned LoRA](https://mlsanity.com/paper/2502.10838/) (2025)
- [Characterizing Continual Learning Scenarios and Strategies for Audio Analysis](https://mlsanity.com/paper/2407.00465/) (2024)
- [Meta-Learning Approaches for Improving Detection of Unseen Speech Deepfakes](https://mlsanity.com/paper/2410.20578/) (2024)
- [Towards Lifelong Learning of Multilingual Text-To-Speech Synthesis](https://mlsanity.com/paper/2110.04482/) (2021)
- [Beyond Silence: Bias Analysis through Loss and Asymmetric Approach in Audio Anti-Spoofing](https://mlsanity.com/paper/2406.17246/) (2024)
- [Audio Deepfake Attribution: An Initial Dataset and Investigation](https://mlsanity.com/paper/2208.10489/) (2022)
- [Exploring Active Data Selection Strategies for Continuous Training in Deepfake Detection](https://mlsanity.com/paper/2502.07269/) (2025)
- [FADEL: Uncertainty-aware Fake Audio Detection with Evidential Deep Learning](https://mlsanity.com/paper/2504.15663/) (2025)
- [TAIL: Text-Audio Incremental Learning](https://mlsanity.com/paper/2503.04258/) (2025)
- [Rehearsal with Auxiliary-Informed Sampling for Audio Deepfake Detection](https://mlsanity.com/paper/2505.24486/) (2025)
- [Traceback Translators Against Forgetting in Continual Fake Speech Detection](https://mlsanity.com/paper/2607.12569/) (2026)
- [Learning as Deepfakes Evolve: RF-Prompt for Continual Audio Deepfake Detection](https://mlsanity.com/paper/2609.37586/) (2026)

## Abstract (arXiv)

Fake audio attack becomes a major threat to the speaker verification system. Although current detection approaches have achieved promising results on dataset-specific scenarios, they encounter difficulties on unseen spoofing data. Fine-tuning and retraining from scratch have been applied to incorporate new data. However, fine-tuning leads to performance degradation on previous data. Retraining takes a lot of time and computation resources. Besides, previous data are unavailable due to privacy in some situations. To solve the above problems, this paper proposes detecting fake without forgetting, a continual-learning-based method, to make the model learn new spoofing attacks incrementally. A knowledge distillation loss is introduced to loss function to preserve the memory of original model. Supposing the distribution of genuine voice is consistent among different scenarios, an extra embedding similarity loss is used as another constraint to further do a positive sample alignment. Experiments are conducted on the ASVspoof2019 dataset. The results show that our proposed method outperforms fine-tuning by the relative reduction of average equal error rate up to 81.62%.