Replay attack detection based on deformable convolutional neural network and temporal-frequency attention model

• Published in: Journal of intelligent systems Vol. 32; no. 1; pp. 588 - 604
• Main Authors: Xie, Dang-en, Hu, Hai-na, Xu, Qiang
• Format: Journal Article
• Language: English
• Published: Berlin De Gruyter 18.07.2023; Walter de Gruyter GmbH
• Subjects: Acoustics; Artificial neural networks; DCNN; Deep learning; Deformation; Discriminant analysis; Energy; Formability; Neural networks; replay attack; Sensors; Social networks; speaker verification; Spectrograms; Speech; Spoofing; temporal-frequency attention model; Time-frequency analysis
• ISSN: 2191-026X; 0334-1860; 2191-026X
• DOI: 10.1515/jisys-2022-0265

As an important identity authentication method, speaker verification (SV) has been widely used in many domains, e.g., mobile financials. At the same time, the existing SV systems are insecure under replay spoofing attacks. Toward a more secure and stable SV system, this article proposes a replay attack detection system based on deformable convolutional neural networks (DCNNs) and a time–frequency double-channel attention model. In DCNN, the positions of elements in the convolutional kernel are not fixed. Instead, they are modified by some trainable variable to help the model extract more useful local information from input spectrograms. Meanwhile, a time–frequency domino double-channel attention model is adopted to extract more effective distinctive features to collect valuable information for distinguishing genuine and replay speeches. Experimental results on ASVspoof 2019 dataset show that the proposed model can detect replay attacks accurately.