Recurrent Neural Networks for Automatic Replay Spoofing Attack Detection

Chen, Zhuxin; Zhang, Wei-Bin; Xie, Zhen; Xu, Xiangmin; Chen, Dongpeng

doi:10.1109/icassp.2018.8462644

Cited by 29 publications

(17 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nonetheless, using the speech frame selection approach [58], the performance of the CQCC-GMM countermeasure improved to 21.60% EER. Replay attack detectors proposed by researchers using Recurrent Neural Networks (RNN) with Filter Bank (Fbank) features have achieved 9.81% EER [16], whereas the one using Deep Neural Networks (DNN) and Support Vector Machine (SVM) classifiers with CQCC and High-Frequency Cepstral Coefficients (HFCC) features have achieved 11.5% EER [82]. The best replay attack detector system in ASVspoof 2017 Challenge hit an EER of 6.73% on the evaluation dataset was a fusion system that adopted several classifiers and features [60].…”

Section: Voice Presentation Attack Detection (Pad)mentioning

confidence: 99%

See 1 more Smart Citation

A survey on presentation attack detection for automatic speaker verification systems: State-of-the-art, taxonomy, issues and future direction

Tan

Hijazi

Khamis

et al. 2021

Multimed Tools Appl

View full text Add to dashboard Cite

The emergence of biometric technology provides enhanced security compared to the traditional identification and authentication techniques that were less efficient and secure. Despite the advantages brought by biometric technology, the existing biometric systems such as Automatic Speaker Verification (ASV) systems are weak against presentation attacks. A presentation attack is a spoofing attack launched to subvert an ASV system to gain access to the system. Though numerous Presentation Attack Detection (PAD) systems were reported in the literature, a systematic survey that describes the current state of research and application is unavailable. This paper presents a systematic analysis of the state-of-the-art voice PAD systems to promote further advancement in this area. The objectives of this paper are two folds: (i) to understand the nature of recent work on PAD systems, and (ii) to identify areas that require additional research. From the survey, a taxonomy of voice PAD and the trend analysis of recent work on PAD systems were built and presented, whereby the recent and relevant articles including articles from Interspeech and ICASSP Conferences, mostly indexed by Scopus, published between 2015 and 2021 were considered. A total of 172 articles were surveyed in this work. The findings of this survey present the limitation of recent works, which include spoof-type dependent PAD. Consequently, the future direction of work on voice PAD for interested researchers is established. The findings of this survey present the limitation of recent works, which include spoof-type dependent PAD. Consequently, the future direction of work on voice PAD for interested researchers is established.

show abstract

Section: Voice Presentation Attack Detection (Pad)mentioning

confidence: 99%

“…Nonetheless, the end-to-end learning approach still contributed more than 15% in overall recent Fig. 16 The proportion of methodology used in voice PAD in recent years Fig. 17 The trend of datasets used in voice PAD in recent years works considered.…”

Section: An Analysis To the Trend Of Voice Pad In Recent Yearsmentioning

confidence: 99%

A survey on presentation attack detection for automatic speaker verification systems: State-of-the-art, taxonomy, issues and future direction

Tan

Hijazi

Khamis

et al. 2021

Multimed Tools Appl

View full text Add to dashboard Cite

show abstract

“…While related studies either proposed individual models for each spoofing method (i.e., PA or LA) [11,12,13], are limited to a specific language (e.g., English), require additional hardware components due to the model's size [14], or are limited to specific applications (impersonation [15,16], replay [17,18,19], speech synthesis [20], or voice conversion [21]), further research is required to advance the accuracy, and universal practicability of such fake voice detection approaches.…”

Section: Introductionmentioning

confidence: 99%

High-Performance Fake Voice Detection on Automatic Speaker Verification Systems for the Prevention of Cyber Fraud with Convolutional Neural Networks

Buettner¹,

Gross²,

Roessler³

et al. 2022

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

This study proposes a highly effective data analytics approach to prevent cyber fraud on automatic speaker verification systems by classifying histograms of genuine and spoofed voice recordings. Our deep learning-based lightweight architecture advances the application of fake voice detection on embedded systems. It sets a new benchmark with a balanced accuracy of 95.64% and an equal error rate of 4.43%, contributing to adopting artificial intelligence technologies in organizational systems and technologies. As fake voice-related fraud causes monetary damage and serious privacy concerns for various applications, our approach improves the security of such services, being of high practical relevance. Furthermore, the post-hoc analysis of our results reveals that our model confirms image texture analysis-related findings of prior studies and discovers further voice signal features (i.e., textural and contextual) that can advance future work in this field.

show abstract

“…In Qian, Chen, Dinkel, and Wu (2017), authors used traditional deep neural networks, convolutional neural networks, and long short‐term memory (LSTM) as deep feature extractors, and LDA as the back‐end classifier. Recurrent neural networks have been used as back‐end classifiers in Chen, Zhang, Xie, Xu, and Chen (2018) and Cai, Cai, Liu, Li, and Li (2017).…”

Section: Introductionmentioning

confidence: 99%

Evolutionary fusion of classifiers trained on linear prediction based features for replay attack detection

Nasersharif

Yazdani

2021

Expert Systems

View full text Add to dashboard Cite

Recently, linear prediction analysis (LP) related features have been successfully used for replay attack detection due to the imperfection in the LP‐based signal produced by recording and playback devices. In this paper, we propose a weighted linear combination of classifier scores for replay attack detection where our classifiers, including Gaussian mixture models (GMMs) and support vector machines (SVMs), are trained on a variety of LP and LP residual‐based features. In this way, we can benefits from all of the LP‐related features when we combine classifiers trained on these features. We determine classifier weights using two evolutionary algorithms: genetic algorithm and particle swarm optimization. Furthermore, we propose a new feature based on performing LP residuals analysis of Mel sub‐band energies. We also propose a deep structure for extracting deep features from LP‐based coefficients to consider the class labels (genuine or spoofed speaker) in the feature extraction process. Results of our classifier system on the ASVspoof 2017 version 2 dataset show equal error rates of 0.3% and 4.8% for its development and evaluation subset, respectively. We also applied our proposed replay attack detection method to an ASV system that has acceptable results.

show abstract

Recurrent Neural Networks for Automatic Replay Spoofing Attack Detection

Cited by 29 publications

References 9 publications

A survey on presentation attack detection for automatic speaker verification systems: State-of-the-art, taxonomy, issues and future direction

A survey on presentation attack detection for automatic speaker verification systems: State-of-the-art, taxonomy, issues and future direction

High-Performance Fake Voice Detection on Automatic Speaker Verification Systems for the Prevention of Cyber Fraud with Convolutional Neural Networks

Evolutionary fusion of classifiers trained on linear prediction based features for replay attack detection

Contact Info

Product

Resources

About