On the Generalization of Fused Systems in Voice Presentation Attack Detection

Gonçalves, Alim P. C.; Violato, Ricardo Paranhos Velloso; Korshunov, Pavel; Marcel, Sébastien; Simões, Flávio

doi:10.23919/biosig.2017.8053516

Cited by 10 publications

(5 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many other features have also been used for replayed speech detection in the context of the ASVspoof 2017 database. Even if the performances of single systems using such features are not always high, they are shown to be complementary when fused at the score level [167], similar to conventional ASV research outside of the spoofing detection. These features include MFCC, IMFCC, rectangular filter cepstral coefficients (RFCCs), PLP, CQCC, spectral centroid magnitude coefficients (SCMC), subband spectral flux coefficient (SSFC), and variable length Teager energy operator energy separation algorithminstantaneous frequency cosine coefficients (VESA-IFCC).…”

Section: Front-ends For Replay Attack Detectionmentioning

confidence: 87%

“…The property of spoofing countermeasures for detecting new kinds of speech presentation attack is an important requirement for their application in the wild. Study explores that countermeasure methods trained with a class of spoofing attacks fail to generalise this for other classes of spoofing attack [191,167]. For example, PAD systems trained with VC and SS based spoofed speech give a very poor performance for playback detection [192].…”

Section: Future Directions Of Anti-spoofing Researchmentioning

confidence: 99%

See 1 more Smart Citation

Introduction to Voice Presentation Attack Detection and Recent Advances

Sahidullah

Delgado

Todisco

et al. 2019

Advances in Computer Vision and Pattern Recognition

View full text Add to dashboard Cite

Over the past few years significant progress has been made in the field of presentation attack detection (PAD) for automatic speaker recognition (ASV). This includes the development of new speech corpora, standard evaluation protocols and advancements in front-end feature extraction and back-end classifiers. The use of standard databases and evaluation protocols has enabled for the first time the meaningful benchmarking of different PAD solutions. This chapter summarises the progress, with a focus on studies completed in the last three years. The article presents a summary of findings and lessons learned from two ASVspoof challenges, the first community-led benchmarking efforts. These show that ASV PAD remains an unsolved problem and that further attention is required to develop generalised

show abstract

Section: Front-ends For Replay Attack Detectionmentioning

confidence: 87%

Section: Future Directions Of Anti-spoofing Researchmentioning

confidence: 99%

Introduction to Voice Presentation Attack Detection and Recent Advances

Sahidullah

Delgado

Todisco

et al. 2019

Advances in Computer Vision and Pattern Recognition

View full text Add to dashboard Cite

show abstract

“…Table 5 compares the performance of our proposed system with other systems trained on the ASVSpoof2017 dataset in terms of EER. When comparing to other models based on GMM, such as System 3 [65], we achieve better performance due to our unique feature set and dataset augmentation. For instance, with the traditional feature set from Experiment 1, a slightly better performance than System 3 is achieved, which uses the same core model, but less features.…”

Section: Comparison With State-of-the-art Systemsmentioning

confidence: 94%

Toward Robust Audio Spoofing Detection: A Detailed Comparison of Traditional and Learned Features

et al. 2019

View full text Add to dashboard Cite

Automatic speaker verification, like every other biometric system, is vulnerable to spoofing attacks. Using only a few minutes of recorded voice of a genuine client of a speaker verification system, attackers can develop a variety of spoofing attacks that might trick such systems. Detecting these attacks using the audio cues present in the recordings is an important challenge. Most existing spoofing detection systems depend on knowing the used spoofing technique. With this research, we aim at overcoming this limitation, by examining robust audio features, both traditional and those learned through an autoencoder, that are generalizable to different types of replay spoofing. Furthermore, we provide a detailed account of all the steps necessary in setting up state-of-the-art audio feature detection, pre-, and postprocessing, such that the (non-audio expert) machine learning researcher can implement such systems. Finally, we evaluate the performance of our robust replay spoofing detection system with a wide variety and different combinations of both extracted and machine learned audio features on the 'out in the wild' ASVspoof 2017 dataset. This dataset contains a variety of new replay spoofing configurations. Since our focus is on examining which features will ensure robustness, we base our system on a traditional Gaussian Mixture Model-Universal Background Model (GMM-UBM). We then systematically investigate the relative contribution of each feature set. The fused models, based on both the known audio features and the machine learned features respectively, have a comparable performance with an Equal Error Rate (EER) of 12. The final best performing model, which obtains an EER of 10.8, is a hybrid system that contains both known and machine learned features, and is trained on an augmented dataset, thus revealing the importance of incorporating both types of features when developing a robust spoofing prediction model.

show abstract

“…In some other efforts, DNN framework is used [24,25,26,27,28]. For replay detection, algorithms were proposed using Electronic Network Frequency (ENF), MFCC and fundamental frequency, linear predictive residual signal, time envelope, stratified scattering decomposition coefficient and Inverse MFCC (IMFCC) respectively [29,30,31,32,33,34,35].…”

Section: Introductionmentioning

confidence: 99%

Detection of Speaker De-identification Disguise Based on Dense Convolutional Network

Wang

Zhu

2020

Preprint

View full text Add to dashboard Cite

Nowadays, speaker disguise is a common operation that presents a great challenge to social security. Therefore, it is important to recognize the authenticity of speech. Most of the current researches focus on speech spoofing, which simulates a target speaker to break through the state-of-art ASV systems by increasing false acceptance rate. Meanwhile, there is another type of disguise, i.e. de-identification, which transforms a speech signal without a target to increase the false rejection rate in order not to be recognized. It has received far less attention. Therefore, in this paper, we investigate the de-identification model and propose a method to detect de-identification speeches from genuine speeches by using a very deep dense convolutional network with 135 layers. The experimental results show that the average accuracy of the proposed method outperforms the reported state-of-the-art methods.

show abstract

On the Generalization of Fused Systems in Voice Presentation Attack Detection

Cited by 10 publications

References 11 publications

Introduction to Voice Presentation Attack Detection and Recent Advances

Introduction to Voice Presentation Attack Detection and Recent Advances

Toward Robust Audio Spoofing Detection: A Detailed Comparison of Traditional and Learned Features

Detection of Speaker De-identification Disguise Based on Dense Convolutional Network

Contact Info

Product

Resources

About