arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2307.01546cs.SDeess.AS
收藏

使用ASR或ASV预训练Conformer用于反欺骗对策

Pretraining Conformer with ASR or ASV for Anti-Spoofing Countermeasure

University of Yamanashi(山梨大学);Duke Kunshan University(昆山杜克大学);Wuhan University(武汉大学)

Yikang Wang, Hiromitsu Nishizaki, Ming Li

AI总结:

本文提出基于迁移学习的MFA-Conformer反欺骗对策,通过ASR或ASV预训练增强鲁棒性,在FAD干净集上EER达0.04%,显著优于基线并与Wav2Vec 2.0方法相当。

摘要:

发现欺骗数据的合成伪影将有助于反欺骗对策(CMs)系统区分欺骗语音和真实语音。Conformer结合了卷积神经网络和Transformer的优点,使其能够聚合全局和局部信息。这可能有利于CM系统捕获隐藏在局部和全局的合成伪影。在本文中,我们提出了基于迁移学习的MFA-Conformer结构用于CM系统。通过使用不同任务预训练Conformer编码器,增强了CM系统的鲁棒性。所提出的方法在中英文欺骗检测数据库上进行了评估。在FAD干净集上,所提出的方法实现了0.04%的等错误率(EER),显著优于基线。我们的系统也与基于Wav2Vec 2.0的预训练方法相当。此外,我们还对不同模型的鲁棒性进行了详细分析。

英文摘要:

Finding synthetic artifacts of spoofing data will help the anti-spoofing countermeasures (CMs) system discriminate between spoofed and real speech. The Conformer combines the best of convolutional neural network and the Transformer, allowing it to aggregate global and local information. This may benefit the CM system to capture the synthetic artifacts hidden both locally and globally. In this paper, we present the transfer learning based MFA-Conformer structure for CM systems. By pre-training the Conformer encoder with different tasks, the robustness of the CM system is enhanced. The proposed method is evaluated on both Chinese and English spoofing detection databases. In the FAD clean set, proposed method achieves an EER of 0.04%, which dramatically outperforms the baseline. Our system is also comparable to the pre-training methods base on Wav2Vec 2.0. Moreover, we also provide a detailed analysis of the robustness of different models.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑