arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2407.03514cs.SDcs.AIeess.AS
收藏

面向音频欺骗检测的基于注意力的对比学习研究

Towards Attention-based Contrastive Learning for Audio Spoof Detection

Reality Defender Inc.(Reality Defender公司)

Chirag Goel, Surya Koppisetti, Ben Colman, Ali Shahriyari, Gaurav Bharaj

AI总结:

本文提出基于注意力的对比学习框架SSAST-CL,利用交叉注意力改进音频欺骗检测中的表示学习,并在ASVSpoof 2021挑战赛上取得有竞争力的性能。

摘要:

视觉Transformer(ViT)在计算机视觉的分类任务中已取得实质性进展。最近,Gong等人(2021)将基于注意力的建模引入若干音频任务。然而,将ViT用于音频欺骗检测任务的研究相对较少。我们弥合这一空白,并将ViT引入该任务。基于微调SSAST(Gong等人,2022)音频ViT模型构建的普通基线取得了次优的等错误率(EER)。为提升性能,我们提出一种新颖的基于注意力的对比学习框架(SSAST-CL),该框架利用交叉注意力辅助表示学习。实验表明,我们的框架成功解耦了真实类和欺骗类,并有助于为该任务学习更好的分类器。在适当的数据增强策略下,基于该框架训练的模型在ASVSpoof 2021挑战赛上取得了具有竞争力的性能。我们提供了对比实验和消融研究以证明上述主张。

英文摘要:

Vision transformers (ViT) have made substantial progress for classification tasks in computer vision. Recently, Gong et. al. '21, introduced attention-based modeling for several audio tasks. However, relatively unexplored is the use of a ViT for audio spoof detection task. We bridge this gap and introduce ViTs for this task. A vanilla baseline built on fine-tuning the SSAST (Gong et. al. '22) audio ViT model achieves sub-optimal equal error rates (EERs). To improve performance, we propose a novel attention-based contrastive learning framework (SSAST-CL) that uses cross-attention to aid the representation learning. Experiments show that our framework successfully disentangles the bonafide and spoof classes and helps learn better classifiers for the task. With appropriate data augmentations policy, a model trained on our framework achieves competitive performance on the ASVSpoof 2021 challenge. We provide comparisons and ablation studies to justify our claim.

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。

↑