Introducing i-vectors for joint anti-spoofing and speaker verification

Khoury, Elie; Kinnunen, Tomi; Сизов, А. С.; Wu, Zhizheng; Marcel, Sébastien

doi:10.21437/interspeech.2014-13

Cited by 20 publications

(10 citation statements)

References 30 publications

(32 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our experiments, the simplest ideas tended to outperform more elaborate ones. For instance, raw power spectrum features and maximum likelihood (ML) trained Gaussian mixture models (GMMs) did a decent job both in detecting both unknown and known attacks, while i-vector [43] based spoofing detection [24,44] yielded much higher error rates.…”

Section: Contribution Of the Present Study: Joint Effect Of Varied At...mentioning

confidence: 99%

Spoofing detection goes noisy: An analysis of synthetic speech detection in the presence of additive noise

Hanili

Kinnunen

Sahidullah

et al. 2016

Speech Communication

View full text Add to dashboard Cite

Automatic speaker verification (ASV) technology is recently finding its way to end-user applications for secure access to personal data, smart services or physical facilities. Similar to other biometric technologies, speaker verification is vulnerable to spoofing attacks where an attacker masquerades as a particular target speaker via impersonation, replay, text-to-speech (TTS) or voice conversion (VC) techniques to gain illegitimate access to the system. We focus on TTS and VC that represent the most flexible, high-end spoofing attacks. Most of the prior studies on synthesized or converted speech detection report their findings using high-quality clean recordings. Meanwhile, the performance of spoofing detectors in the presence of additive noise, an important consideration in practical ASV implementations, remains largely unknown. To this end, our study provides a comparative analysis of existing state-of-the-art, off-the-shelf synthetic speech detectors under additive noise contamination with a special focus on front-end processing that has been found critical. Our comparison includes eight acoustic feature sets, five related to spectral magnitude and three to spectral phase information. All the methods contain a number of internal control parameters. Except for feature post-processing steps (deltas and cepstral mean normalization) that we optimized for each method, we fix the internal control parameters to their default values based on literature, and compare all the variants using the exact same dimensionality and back-end system. In addition to the eight feature sets, we consider two alternative classifier back-ends: Gaussian mixture model (GMM) and i-vector, the latter with both cosine scoring and probabilistic linear discriminant analysis (PLDA) scoring. Our extensive analysis on the recent ASVspoof 2015 challenge provides new insights to the robustness of the spoofing detectors. Firstly, unlike in most other speech processing tasks, all the compared spoofing detectors break down even at relatively high signal-to-noise ratios (SNRs) and fail to generalize to noisy conditions even if performing excellently on clean data. This indicates both difficulty of the task, as well as potential to over-fit the methods easily. Secondly, speech enhancement pre-processing is not found helpful. Thirdly, GMM back-end generally outperforms the more involved i-vector back-end. Fourthly, concerning the compared features, the Mel-frequency cepstral coefficient (MFCC) and subband spectral centroid magnitude coefficient (SCMC) features perform the best on average though the winner method depends on SNR and noise type. Finally, a study with two score fusion strategies shows that combining different feature based systems improves recognition accuracy for known and unknown attacks in both clean and noisy conditions. In particular, simple score averaging fusion, as opposed to weighted fusion with logistic loss weight optimization, was found to work better, on average. For clean speech, it provides 88% and 28% relative improvements ...

show abstract

Section: Contribution Of the Present Study: Joint Effect Of Varied At...mentioning

confidence: 99%

Spoofing detection goes noisy: An analysis of synthetic speech detection in the presence of additive noise

Hanili

Kinnunen

Sahidullah

et al. 2016

Speech Communication

View full text Add to dashboard Cite

show abstract

“…Previous models include the ever so popular i-vector [3,4,5], which when used as a standalone model does not perform arXiv:2007.13060v1 [eess.AS] 26 Jul 2020 well. Another popular approach is to use the traditional GMM model.…”

Section: Modelmentioning

confidence: 99%

End-to-end spoofing detection with raw waveform CLDNNS

Dinkel

Chen

et al. 2017

2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Albeit recent progress in speaker verification generates powerful models, malicious attacks in the form of spoofed speech, are generally not coped with. Recent results in ASVSpoof2015 and BTAS2016 challenges indicate that spoof-aware features are a possible solution to this problem. Most successful methods in both challenges focus on spoof-aware features, rather than focusing on a powerful classifier. In this paper we present a novel raw waveform based deep model for spoofing detection, which jointly acts as a feature extractor and classifier, thus allowing it to directly classify speech signals. This approach can be considered as an end-to-end classifier, which removes the need for any pre-or post-processing on the data, making training and evaluation a streamlined process, consuming less time than other neural-network based approaches. The experiments on the BTAS2016 dataset show that the system performance is significantly improved by the proposed raw waveform convolutional long short term neural network (CLDNN), from the previous best published 1.26% half total error rate (HTER) to the current 0.82% HTER. Moreover it shows that the proposed system also performs well under the unknown (RE-PH2-PH3,RE-LPPH2-PH3) conditions.

show abstract

“…Research work on anti-spoofing can be divided into one of the three categories: Feature Learning [5,6,7,8,9,10,11,12,13,14], Statistical Modeling [4,15,16,17], and Deep Neural Network (DNN) [18,19,20,21,22,23,24,25]. Having witnessed the successes of DNNs in ASVspoof 2017, we decided to explore and extend several DNN-based systems for the ASVspoof 2019 Challenge.…”

Section: Introductionmentioning

confidence: 99%

ASSERT: Anti-Spoofing with Squeeze-Excitation and Residual Networks

et al. 2019

View full text Add to dashboard Cite

We present JHU's system submission to the ASVspoof 2019 Challenge: Anti-Spoofing with Squeeze-Excitation and Residual neTworks (ASSERT). Anti-spoofing has gathered more and more attention since the inauguration of the ASVspoof Challenges, and ASVspoof 2019 dedicates to address attacks from all three major types: text-to-speech, voice conversion, and replay. Built upon previous research work on Deep Neural Network (DNN), ASSERT is a pipeline for DNN-based approach to anti-spoofing. ASSERT has four components: feature engineering, DNN models, network optimization and system combination, where the DNN models are variants of squeeze-excitation and residual networks. We conducted an ablation study of the effectiveness of each component on the ASVspoof 2019 corpus, and experimental results showed that ASSERT obtained more than 93% and 17% relative improvements over the baseline systems in the two sub-challenges in ASVspooof 2019, ranking ASSERT one of the top performing systems. Code and pretrained models will be made publicly available.

show abstract

Introducing i-vectors for joint anti-spoofing and speaker verification

Cited by 20 publications

References 30 publications

Spoofing detection goes noisy: An analysis of synthetic speech detection in the presence of additive noise

Spoofing detection goes noisy: An analysis of synthetic speech detection in the presence of additive noise

End-to-end spoofing detection with raw waveform CLDNNS

ASSERT: Anti-Spoofing with Squeeze-Excitation and Residual Networks

Contact Info

Product

Resources

About