Learning From Yourself: A Self-distillation Method For Fake Speech Detection

Abstract

In this paper, we propose a novel self-distillation method for fake speech detection (FSD), which can significantly improve the performance of FSD without increasing the model complexity. For FSD, some fine-grained information is very important, such as spectrogram defects, mute segments, and so on, which are often perceived by shallow networks. However, shallow networks have much noise, which can not capture this very well. To address this problem, we propose using the deepest network instruct shallow network for enhancing shallow networks. Specifically, the networks of FSD are divided into several segments, the deepest network being used as the teacher model, and all shallow networks become multiple student models by adding classifiers. Meanwhile, the distillation path between the deepest network feature and shallow network features is used to reduce the feature difference. A series of experimental results on the ASVspoof 2019 LA and PA datasets show the effectiveness of the proposed

Stats

• citations27
• S2 citations—
• github stars0
• HF likes0
• heat score10.85
• arxiv keyxue2023learning

Related papers

• Continual Learning For Fake Audio Detection (2021)11.49
• Spatial Reconstructed Local Attention Res2net With F0 Subband For Fake Speech Detection (2023)8.82
• Mixture Of Experts Fusion For Fake Audio Detection Using Frozen Wav2vec 2.0 (2024)0.00
• Dual-branch Knowledge Distillation For Noise-robust Synthetic Speech Detection (2023)9.07
• FADEL: Uncertainty-aware Fake Audio Detection With Evidential Deep Learning (2025)0.00
• Deep Residual Neural Networks For Audio Spoofing Detection (2019)0.00
• DIN-CTS: Low-complexity Depthwise-inception Neural Network With Contrastive Training Strategy For Deepfake Speech Detection (2025)2.26
• Multi-perspective Information Fusion Res2net With Randomspecmix For Fake Speech Detection (2023)0.00