Dynamically Mitigating Data Discrepancy With Balanced Focal Loss For Replay Attack Detection

Abstract

It becomes urgent to design effective anti-spoofing algorithms for vulnerable automatic speaker verification systems due to the advancement of high-quality playback devices. Current studies mainly treat anti-spoofing as a binary classification problem between bonafide and spoofed utterances, while lack of indistinguishable samples makes it difficult to train a robust spoofing detector. In this paper, we argue that for anti-spoofing, it needs more attention for indistinguishable samples over easily-classified ones in the modeling process, to make correct discrimination a top priority. Therefore, to mitigate the data discrepancy between training and inference, we propose D3M, to leverage a balanced focal loss function as the training objective to dynamically scale the loss based on the traits of the sample itself. Besides, in the experiments, we select three kinds of features that contain both magnitude-based and phase-based information to form complementary and informative features. Exp

Stats

• citations2
• S2 citations—
• github stars30
• HF likes0
• heat score6.56
• arxiv keydou2020dynamically

Related papers

• Toward Improving Synthetic Audio Spoofing Detection Robustness Via Meta-learning And Disentangled Training With Adversarial Examples (2024)6.77
• Continual Learning For Fake Audio Detection (2021)11.49
• The DKU Replay Detection System For The Asvspoof 2019 Challenge: On Data Augmentation, Feature Representation, Classification, And Fusion (2019)12.25
• Audio-replay Attack Detection Countermeasures (2017)6.34
• Transforming Acoustic Characteristics To Deceive Playback Spoofing Countermeasures Of Speaker Verification Systems (2018)6.34
• Deep Generative Variational Autoencoding For Replay Spoof Detection In Automatic Speaker Verification (2020)9.76
• Optimizing A-dcf For Spoofing-robust Speaker Verification (2024)3.58
• Self-attention And Hybrid Features For Replay And Deep-fake Audio Detection (2024)0.00