# Dual-Branch Knowledge Distillation for Noise-Robust Synthetic Speech Detection

arXiv: https://arxiv.org/abs/2310.08869 · HTML: https://mlsanity.com/paper/2310.08869/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Speech and audio distillation, one of 12 directions in Knowledge distillation.
- Builds on 37 of its 72 references that are in the map.
- First posted on arXiv 13 Oct 2023.

## Placed in

- Knowledge distillation / [Speech and audio distillation](https://mlsanity.com/distill/speech-and-audio-distillation/)

## Builds on (most cited first, 30 of 37)

- [Learning without Forgetting](https://mlsanity.com/paper/1606.09282/) (2016)
- [Knowledge Distillation: A Survey](https://mlsanity.com/paper/2006.05525/) (2020)
- [SphereFace: Deep Hypersphere Embedding for Face Recognition](https://mlsanity.com/paper/1704.08063/) (2017)
- [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://mlsanity.com/paper/2111.09296/) (2021)
- [Improved Knowledge Distillation via Teacher Assistant](https://mlsanity.com/paper/1902.03393/) (2019)
- [Speech Enhancement and Dereverberation with Diffusion-based Generative Models](https://mlsanity.com/paper/2208.05830/) (2022)
- [A Survey on Neural Speech Synthesis](https://mlsanity.com/paper/2106.15561/) (2021)
- [AASIST: Audio Anti-Spoofing using Integrated Spectro-Temporal Graph Attention Networks](https://mlsanity.com/paper/2110.01200/) (2021)
- [ASVspoof 2019: Future Horizons in Spoofed and Fake Audio Detection](https://mlsanity.com/paper/1904.05441/) (2019)
- [End-to-end anti-spoofing with RawNet2](https://mlsanity.com/paper/2011.01108/) (2020)
- [ASVspoof 2021: accelerating progress in spoofed and deepfake speech detection](https://mlsanity.com/paper/2109.00537/) (2021)
- [ADD 2022: the First Audio Deep Synthesis Detection Challenge](https://mlsanity.com/paper/2202.08433/) (2022)
- [End-to-End Spectro-Temporal Graph Attention Networks for Speaker Verification Anti-Spoofing and Speech Deepfake Detection](https://mlsanity.com/paper/2107.12710/) (2021)
- [Investigating self-supervised front ends for speech spoofing countermeasures](https://mlsanity.com/paper/2111.07725/) (2021)
- [FReTAL: Generalizing Deepfake Detection using Knowledge Distillation and Representation Learning](https://mlsanity.com/paper/2105.13617/) (2021)
- [Replay and Synthetic Speech Detection with Res2net Architecture](https://mlsanity.com/paper/2010.15006/) (2020)
- [Towards End-to-End Synthetic Speech Detection](https://mlsanity.com/paper/2106.06341/) (2021)
- [Speech is Silver, Silence is Golden: What do ASVspoof-trained Models Really Learn?](https://mlsanity.com/paper/2106.12914/) (2021)
- [How Bad Are Artifacts?: Analyzing the Impact of Speech Enhancement Errors on ASR](https://mlsanity.com/paper/2201.06685/) (2022)
- [A Noise-Robust Self-supervised Pre-training Model Based Speech Representation Learning for Automatic Speech Recognition](https://mlsanity.com/paper/2201.08930/) (2022)
- [Improving noise robust automatic speech recognition with single-channel time-domain enhancement network](https://mlsanity.com/paper/2003.03998/) (2020)
- [Interactive Feature Fusion for End-to-End Noise-Robust Speech Recognition](https://mlsanity.com/paper/2110.05267/) (2021)
- [A Study On Data Augmentation In Voice Anti-Spoofing](https://mlsanity.com/paper/2110.10491/) (2021)
- [Wav2vec-Switch: Contrastive Learning from Original-noisy Speech Pairs for Robust Speech Recognition](https://mlsanity.com/paper/2110.04934/) (2021)
- [Learning From Yourself: A Self-Distillation Method for Fake Speech Detection](https://mlsanity.com/paper/2303.01211/) (2023)
- [VoiceID Loss: Speech Enhancement for Speaker Verification](https://mlsanity.com/paper/1904.03601/) (2019)
- [Multitask-Based Joint Learning Approach To Robust ASR For Radio Communication Speech](https://mlsanity.com/paper/2107.10701/) (2021)
- [An Empirical Study on Channel Effects for Synthetic Voice Spoofing Countermeasure Systems](https://mlsanity.com/paper/2104.01320/) (2021)
- [Dual Application of Speech Enhancement for Automatic Speech Recognition](https://mlsanity.com/paper/2011.03840/) (2020)
- [Learning to Enhance or Not: Neural Network-Based Switching of Enhanced and Observed Signals for Overlapping Speech Recognition](https://mlsanity.com/paper/2201.03881/) (2022)

## Built on by (most cited first, 5 of 5)

- [Mixture of Experts Fusion for Fake Audio Detection Using Frozen wav2vec 2.0](https://mlsanity.com/paper/2409.11909/) (2024)
- [BiCrossMamba-ST: Speech Deepfake Detection with Bidirectional Mamba Spectro-Temporal Cross-Attention](https://mlsanity.com/paper/2505.13930/) (2025)
- [Audio-Visual Representation Learning via Knowledge Distillation from Speech Foundation Models](https://mlsanity.com/paper/2502.05766/) (2025)
- [Frequency-mix Knowledge Distillation for Fake Speech Detection](https://mlsanity.com/paper/2406.09664/) (2024)
- [Time-Frequency Consistency Learning for Robust Speech Deepfake Detection](https://mlsanity.com/paper/2607.17761/) (2026)

## Abstract (arXiv)

Most research in synthetic speech detection (SSD) focuses on improving performance on standard noise-free datasets. However, in actual situations, noise interference is usually present, causing significant performance degradation in SSD systems. To improve noise robustness, this paper proposes a dual-branch knowledge distillation synthetic speech detection (DKDSSD) method. Specifically, a parallel data flow of the clean teacher branch and the noisy student branch is designed, and interactive fusion module and response-based teacher-student paradigms are proposed to guide the training of noisy data from both the data distribution and decision-making perspectives. In the noisy student branch, speech enhancement is introduced initially for denoising, aiming to reduce the interference of strong noise. The proposed interactive fusion combines denoised features and noisy features to mitigate the impact of speech distortion and ensure consistency with the data distribution of the clean branch. The teacher-student paradigm maps the student's decision space to the teacher's decision space, enabling noisy speech to behave similarly to clean speech. Additionally, a joint training method is employed to optimize both branches for achieving global optimality. Experimental results based on multiple datasets demonstrate that the proposed method performs effectively in noisy environments and maintains its performance in cross-dataset experiments. Source code is available at https://github.com/fchest/DKDSSD.