Pith. sign in

REVIEW 3 major objections 6 minor 2 cited by

Tandem spoofing-robust automatic speaker verification based on time-domain embeddings

T0 review · 3 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Gender-dependent countermeasures built from waveform-amplitude embeddings reduce tandem spoofing cost in speaker verification.

desk verdict A candid system study whose central GD-vs-GI claim rests on evaluation-set threshold tuning; the development set contradicts it, but the fusion analysis is honest and the work deserves a rigorous referee. read the letter →

arxiv 2412.17133 v1 pith:PLMQCXVW submitted 2024-12-22 eess.AS

Avishai Weizman , Yehuda Ben-Shimol , Itshak Lapidot This is my paper · ORCID
classification eess.AS
keywords AutomaticspeakerverificationAnti-spoofingCountermeasuresystemTime-domainembeddingsProbabilitymassfunctionGenderrecognitionTandemdetectioncostLogicalaccessattacks
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that a simple, explainable signal representation—the distribution of waveform amplitudes, measured per filter-bank channel and compared against gender- and class-specific reference distributions—can drive both gender recognition and spoofing detection. The authors further claim that splitting the countermeasure by gender improves a tandem speaker-verification system. On ASVspoof2019 logical-access attacks, the gender-dependent countermeasure reaches male and female equal error rates of 8.67% and 10.12%, and the full tandem system achieves a lower normalized minimum t-DCF than the gender-independent version (0.2709 versus 0.3178 on the evaluation set). In matched conditions the countermeasure is near-perfect, so the interesting claim is that gender separation helps when unseen attacks arrive. This matters because most countermeasures rely on frequency-domain or self-supervised features, whereas this route works directly on raw waveform statistics.

What carries the argument

The PMF time embedding: each input waveform passes through 10 Gammatone and 10 inverse Gammatone filters (20 channels), each channel's amplitude histogram is normalized to a probability mass function with $2^{16}$ bins, and eight similarity measures (including Hellinger distance, Kullback-Leibler divergence, and normalized cross-correlation) are computed between the input channel PMF and reference PMFs of training groups such as genuine/spoofed or male/female. Each embedding component is the difference between similarity to one class and similarity to the other class, giving $20 \times 8 = 160$ numbers. The countermeasure splits these 160 numbers into 16 groups by similarity metric and filter type, feeds each group to a small fully connected network with a one-class softmax loss, and trains male and female systems separately. A gradient-boosted tree classifier on the same embeddings performs the gender routing.

What would settle it

Re-run the same gender-dependent and gender-independent systems on a later logical-access benchmark with unseen text-to-speech and voice-conversion attacks: if the gender-dependent normalized minimum t-DCF is not below the gender-independent value with non-overlapping bootstrap confidence intervals, or if shuffling gender labels leaves evaluation EERs unchanged, the paper's central comparison fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that a 160-dimensional embedding built from probability mass functions of waveform amplitudes—20 filtered channels, eight similarity measures, and one subtraction per channel between similarity to a spoofed-class reference and similarity to a genuine-class reference—carries enough speaker-gender and genuineness information to serve as the input to a spoofing-robust speaker verification system. Using these embeddings, the paper shows that countermeasures trained separately for male and female speech outperform a gender-independent countermeasure on the ASVspoof2019 evaluation set, both in countermeasure EER (combined gender-dependent 9.68% versus gender-independent 10.21%) and in normalized minimum t-DCF for the tandem countermeasure-plus-verifier pipeline (gender-dependent 0.2709 versus gender-independent 0.3178 with estimated gender). A gender-recognition front-end built on the same embeddings misclassifies 0.94% of male and 1.79% of female utterances. Fusing the time-embedding scores with conventional LFCC-based countermeasures helps when fusion parameters are tuned on the evaluation set, but tuning them on the development set degrades evaluation performance, exposing the method's sensitivity to distribution shift.

Load-bearing premise

The PMF reference models built from the training set's male, female, genuine, and spoofed speech must stay representative for the unseen attack algorithms in the evaluation set; if they do not, the embedding scores stop separating spoofed from genuine and the gender advantage disappears.

Editorial extensions

If this is right

  • With estimated gender labels, the gender-dependent tandem system lowers normalized minimum t-DCF from 0.3178 (gender-independent) to 0.2709 on the evaluation set.
  • Per-gender countermeasure EERs improve over the earlier PMF baseline: male 8.67% versus 12.09%, and female 10.12% versus 12.99%.
  • Gender routing costs little: the classifier on the same embeddings misclassifies only 0.94% of male and 1.79% of female evaluation utterances.
  • Fusing time-embedding scores with LFCC-based countermeasures lowers evaluation EER relative to either system alone when fusion hyper-parameters are tuned on the evaluation set, e.g., OCSoftmax plus gender-dependent time embeddings reaches 1.78% female EER versus 2.05% for the LFCC-only system.
  • If fusion hyper-parameters are tuned on the development set, the fusion no longer helps on the evaluation set, so the fusion benefit is contingent on matching the fusion stage to evaluation conditions.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • An implicit consequence is that per-gender thresholding, not only per-gender models, is doing part of the work; the paper's own threshold maps show the EER operating point is not optimal, so a gender-aware threshold policy alone might recover part of the gain at much lower cost.
  • Because the paper's visualization shows genuine speech staying in place while attacks shift between development and evaluation, test-time clustering of the embedding space could allow reference PMFs to be updated without retraining the countermeasure.
  • The evaluation evidence is based on one benchmark with a small speaker set, so whether gender separation still helps with more training speakers and newer attack algorithms, such as later challenge editions, remains an open question.
  • The sensitivity of fusion to hyper-parameter selection suggests that simple linear score fusion discards complementary information; a small domain-adapted fusion layer could be more robust to the dev-to-eval shift.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The paper proposes a tandem spoofing-robust automatic speaker verification (SASV) system evaluated on the ASVspoof2019 logical access database. The central novelty is a countermeasure (CM) based on time-domain embeddings derived from the probability mass function (PMF) of filtered waveform amplitudes, including a gender-recognition front-end built from the same embeddings. The system combines gender-dependent CMs with ECAPA-TDNN speaker verification, and it also fuses the proposed CM scores with LFCC-based ResNet CM scores. The headline claims are that the gender-dependent (GD) architecture outperforms a gender-independent (GI) architecture on the evaluation set and that the approach improves generalization in SASV, as measured by EER, normalized min t-DCF, and normalized min a-DCF.

Significance. If the empirical comparison were conducted under a fixed, test-set-free protocol, the paper would make a useful contribution: the time-domain PMF embedding is compact, interpretable, and appears to carry both spoofing and gender information, and the authors provide bootstrap confidence intervals and compare against their own prior time-embedding baselines. The introduction of gender-conditioned operating points in a tandem SASV system is a reasonable research direction. However, as reported, the central GD-versus-GI and fusion-generalization claims are not supported because several load-bearing thresholds and hyperparameters are selected on the evaluation set itself, while the development set points in the opposite direction. The paper needs a re-analysis under a development-only tuning protocol, or a substantially weakened statement of the conclusions.

major comments (3)
  1. [Section III.C and Section V.C, Tables II, V, VI, VII, VIII] The GD-versus-GI comparison is confounded by evaluation-set threshold fitting. Section III.C states that ASV detection thresholds are fixed according to the EER threshold for each gender on the evaluation set, Section V.C states that CM thresholds are determined for each database subset, and Figure 7 optimizes both thresholds on the evaluation set. GD therefore receives two gender-specific operating points chosen on the test set, while GI receives one, which is an extra degree of freedom that can only lower the measured GD error. The development set contradicts the reported advantage: Table VII shows Dev. GD min t-DCF 0.0039 versus Dev. GI 0.0016, Table VIII shows Dev. GD min a-DCF 0.0064 versus Dev. GI 0.0046, and Table II shows Dev. GD EER 0.26% versus Dev. GI 0.10%. Additionally, the male CM and the female/GI CM use different network architectures and losses, so GD versus GI does not isolate gender dependence. The authors should report a development-set-only protocol with all thresholds fixed before touching the evaluation set, or explicitly label the current results as post hoc operating-point selection and remove the generalization claim.
  2. [Section V.E, Tables X, XI, XII] The fusion results also rely on evaluation-set fitting. Equation (14) uses a fusion weight alpha whose optimal value is estimated on the evaluation set (Table X), and the classifier-based fusion in Tables XI and XII includes grid searches conducted on the evaluation set. With development-set-based alpha, the evaluation EERs are worse than the LFCC-alone baselines in several cases (e.g., Table X Softmax+GD 5.30% versus Softmax 5.03%, and Table XI OCSoftmax+GD 2.62% versus OCSoftmax 2.18%). The abstract and Section V.D claim improved generalization, but the reported fusion gains are not obtained under a test-set-free protocol. The authors should either present a genuinely held-out fusion protocol or rephrase the contribution as a sensitivity analysis showing how evaluation-set tuning inflates apparent fusion performance.
  3. [Section V.B, Section IV.A, Section VI] The generalization claim is further weakened by the evidence inside the paper itself. Section V.B's UMAP analysis shows no clear genuine/spoofed separation on the evaluation set, while Section IV.A notes that two evaluation attacks share algorithms with training attacks, so the evaluation set is not a pure test of generalization to entirely novel synthesis/conversion algorithms. The conclusion in Section VI acknowledges significant degradation in unmatched conditions and overlapping confidence intervals for the fusion results. The claim that the GD system demonstrates 'improved generalization ability' therefore goes beyond what the reported experimental design can establish. The paper should either provide a development-set-only evaluation of the GD/GI comparison or explicitly restrict the conclusions to post hoc performance on the ASVspoof2019 evaluation set.
minor comments (6)
  1. [Section V.B, Figure 3 caption] The phrase 'gender dependet' should read 'gender dependent'.
  2. [Section IV.B and Section V.E] The symbol alpha is used both for the confidence level in the bootstrap procedure ('alpha = 5') and for the fusion weight in Equation (14); please disambiguate these two uses, since the confidence level is presumably 5% rather than 5.
  3. [Table VIII] The Dev. GD row reports a value of 0.0064 with confidence interval [0.0102, 0.0138], which does not contain the point estimate and appears to be a formatting or copying error; please check the interval computation.
  4. [Table I] Some confidence intervals in the Random Forest row, such as [0.02, 0.48] for a point estimate of 1.52 and [0.00, 0.04] for a point estimate of 1.97, look implausibly narrow or misplaced; please verify the bootstrap output.
  5. [Section V.E] The text says the LFCC-based systems were adopted from [63], but [63] is the original ResNet paper; the actual source appears to be [29] or another system description, so the citation should be corrected.
  6. [Figure 7] The text refers to cyan and red operating points on the normalized unconstrained t-DCF surfaces, but the figure caption does not explain these markers; please add a legend or caption explanation.

Circularity Check

2 steps flagged · score 6.0 of 10

Eval-set per-gender thresholds and fusion alpha are fitted to the test set, so the reported GD-over-GI and fusion gains are not out-of-sample predictions; the development set already contradicts the GD advantage.

  1. fitted input called prediction [Section III.C and Section V.C-V.D (Tables V-VII, Figure 7)]
    "The ASV systems’ detection thresholds were fixed according to the EER threshold for each gender on the evaluation set, as described in [45]. ... In contrast, in the CM GD scenario, two thresholds were utilized, one for each gender. ... The results demonstrate that the GD system architecture outperforms the GI system for the evaluation set, aligning with findings in [25]."

    The central claim is that GD generalizes better than GI, but GD is given gender-specific ASV and CM thresholds chosen on the evaluation set while GI is given a single threshold. The eval-only advantage therefore reflects extra threshold degrees of freedom fitted to the test set, not an out-of-sample property. Table VII shows Dev. GD min t-DCF 0.0039 versus Dev. GI 0.0016, so the development set already contradicts the claimed advantage; the GD edge appears precisely on the set where its extra operating points were tuned.

  2. fitted input called prediction [Section V.E, Eq. (14), Tables X-XII]
    "Consequently, the minimum EER was also evaluated by estimating the optimal α value on the evaluation set. ... It can be seen that the optimal α value, selected based on the evaluation set, resulted in enhanced performance."

    The fusion weight α is fitted by minimizing EER on the evaluation set and the resulting improvement is then reported as evidence for the fusion approach. Since the evaluation set is also the set on which the fused result is measured, the reported gain is partly an artifact of test-set optimization rather than a prediction of generalization; the paper itself notes that dev-selected α degrades evaluation performance.

full rationale

The core PMF time-embedding derivation (Eqs. (1)-(2)) is not circular: the PMF models are computed from the training set, the embeddings are fixed functions of those models, and the classifier results are measured on the held-out ASVspoof2019 evaluation set. No load-bearing uniqueness theorem is imported from the authors' prior work, and comparisons against [25], [34] are legitimate external/prior baselines. The circularity is confined to the operating-point and fusion-selection stages. The paper selects per-gender ASV thresholds on the evaluation set (Section III.C), determines per-gender CM thresholds per database subset (Section V.C), and chooses the fusion weight α by minimizing EER on the evaluation set (Section V.E). It then reports the resulting evaluation-set GD and fusion numbers as demonstrating improved generalization. That is a fitted input reported as a prediction: the improvement is partly forced by test-set tuning, and the development-set results (e.g., Table VII: GD 0.0039 vs GI 0.0016) point in the opposite direction. The feature-extraction and CM-training content itself remains independently testable, so this is partial circularity, not a fully definitional collapse.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central empirical results depend on a handful of tunable quantities: decision thresholds set per subset, fusion weight alpha, CM network hyperparameters, and the filterbank/similarity configuration inherited from prior work. The axioms are standard external benchmark definitions plus the domain assumption that training-set PMF models transfer to unseen attacks.

free parameters (4)
  • ASV and CM decision thresholds = ASV EER thresholds on eval: 0.352 (male), 0.401 (female); CM thresholds per subset
    Section III.C sets ASV thresholds from EER on the evaluation set; Section V.C sets CM thresholds per subset. All EER, minDCF, t-DCF, and a-DCF numbers depend on these choices.
  • Fusion weight alpha (Eq. 14) = alpha_Dev 0.36-0.83; alpha_Eval 0.01-0.05 (Table X)
    Weighted average of GD time-embedding and LFCC scores. The optimal alpha changes drastically between dev and eval, and drives whether fusion appears to help.
  • CM network architecture hyperparameters = Male FC 5-40-1; female/GI FC 10-40/80-48/32; dropout 0.2; batch 256/32/128; epochs 300/100/200
    Section V.B says these were found by exhaustive grid search on the development set. They define the CM systems whose EERs are reported.
  • Filterbank and similarity configuration = 20 filters (10 Gammatone + 10 Inverse Gammatone), 8 similarity measures, 2^16 bins
    Section II follows [25], [36] without ablation in this paper; the 160-dimensional embedding and all subsequent results depend on these choices.
assumptions (4)
  • domain assumption Empirical PMFs estimated on the training-set groups are representative of evaluation-set genuine and spoofed speech.
    Section II builds class PMF models from training data; Section V.B shows poor genuine/spoofed separation on the evaluation set, so the method rests on transfer that is not directly validated.
  • ad hoc to paper The eight selected similarity measures adequately capture PMF differences for both spoofing and gender.
    Section II adopts Quadratic Chi, NCC, Hellinger, Intersection, KL, KLS, JSD, and modified KS from [25] without a task-specific derivation.
  • domain assumption ECAPA-TDNN trained on VoxCeleb serves as a valid fixed ASV backend under the ASVspoof2019 LA protocol.
    Section III.C uses SpeechBrain ECAPA-TDNN with no domain adaptation, only per-gender EER thresholds.
  • standard math t-DCF and a-DCF cost parameters and normalization from [48] and [49] apply unchanged.
    Section IV.B defines normalized t-DCF and a-DCF using the challenge's standard formulas and priors.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Tandem spoofing-robust automatic speaker verification based on time-domain embeddings." pith.science (2026). https://pith.science/paper/PLMQCXVW

@misc{pith2026241217133,
  author       = {Pith},
  title        = {Pith review of: Tandem spoofing-robust automatic speaker verification based on time-domain embeddings},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PLMQCXVW}},
  note         = {Machine review of arXiv:2412.17133}
}
read the original abstract

Spoofing-robust automatic speaker verification (SASV) systems are a crucial technology for the protection against spoofed speech. In this study, we focus on logical access attacks and introduce a novel approach to SASV tasks. A novel representation of genuine and spoofed speech is employed, based on the probability mass function (PMF) of waveform amplitudes in the time domain. This methodology generates novel time embeddings derived from the PMF of selected groups within the training set. This paper highlights the role of gender segregation and its positive impact on performance. We propose a countermeasure (CM) system that employs time-domain embeddings derived from the PMF of spoofed and genuine speech, as well as gender recognition based on male and female time-based embeddings. The method exhibits notable gender recognition capabilities, with mismatch rates of 0.94% and 1.79% for males and females, respectively. The male and female CM systems achieve an equal error rate (EER) of 8.67% and 10.12%, respectively. By integrating this approach with traditional speaker verification systems, we demonstrate improved generalization ability and tandem detection cost function evaluation using the ASVspoof2019 challenge database. Furthermore, we investigate the impact of fusing the time embedding approach with traditional CM and illustrate how this fusion enhances generalization in SASV architectures.

Figures

Figures reproduced from arXiv: 2412.17133 by the authors.

Figure 1
Figure 1. Time embedding generation process [34]. predefined group may be based on gender or the distinction between spoofed and genuine speech. An additional example is the generation of models comprising more than two categories: male spoofed, female spoofed, genuine male, and genuine female speech. In this work, we followed the approach in [25], [36]. Each speech signal is subjected to processing through 10 Gammatone and 1… view at source ↗
Figure 2
Figure 2. The Spoofing-robust speaker verification architecture. The same audio signal, is fed into the time embedding generation system (on the left) and [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Neural networks structures: left side for the male gender and the [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: 3 dimensional PCA projection of time embeddings grouped by [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: UMAP dimensionality reduction of time embeddings trained on [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Histogram and kernel density estimation (KDE) of attack analysis for ASVSpoof2019 on the ECAPA-TDNN system Table IV ECAPA-TDNN PERFORMANCE ON THE ASVSPOOF2019 LA DATABASE IN TERMS OF MINDCF AND EER. THE THRESHOLD IS DETERMINED BASED ON THE EER OF EACH DATABASE SUBSET, …
Figure 7
Figure 7. Figure 7: The minimum normalized unconstrained t-DCF : left - males, [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: Scores of GD CM System vs. OCSoftmax CM System. [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. ATMM-SAGA: Alternating Training for Multi-Module with Score-Aware Gated Attention SASV system

    eess.AS 2025-05 conditional novelty 5.0 of 10

    A gated fusion of speaker embeddings and spoofing scores, trained with alternating weight freezing, achieves 2.18% SASV-EER on ASVspoof2019 LA evaluation.

  2. Unmasking Synthetic Realities in Generative AI: A Comprehensive Review of Adversarially Robust Deepfake Detection Systems

    cs.CR 2025-07 conditional novelty 3.0 of 10

    A systematic review of deepfake detection finds a pervasive lack of adversarial robustness evaluation across all modalities and calls for resilient, modality-agnostic detectors.

Reference graph

Works this paper leans on

64 extracted references · 50 canonical work pages · cited by 2 Pith papers

  1. [34]

    Spoofing-robust speaker verification based on time-domain embedding,

    A. Weizman, Y . Ben-Shimol, and I. Lapidot, “Spoofing-robust speaker verification based on time-domain embedding,” in Cyber Security, Cryp- tology, and Machine Learning , S. Dolev, M. Elhadad, M. Kutyłowski, and G. Persiano, Eds. Cham: Springer Nature Switzerland, 2025, pp. 64–78

    work page 2025

  2. [1]

    50 years of biomet- ric research: Accomplishments, challenges, and opportunities,

    A. K. Jain, K. Nandakumar, and A. Ross, “50 years of biomet- ric research: Accomplishments, challenges, and opportunities,” Pattern recognition letters, vol. 79, pp. 80–105, 2016

    work page 2016

  3. [2]

    Biometrics systems under spoofing attack: an evaluation methodology and lessons learned,

    A. Hadid, N. Evans, S. Marcel, and J. Fierrez, “Biometrics systems under spoofing attack: an evaluation methodology and lessons learned,” IEEE Signal Processing Magazine , vol. 32, no. 5, pp. 20–30, 2015

    work page 2015

  4. [3]

    Biometrics: a tool for information security,

    A. K. Jain, A. Ross, and S. Pankanti, “Biometrics: a tool for information security,” IEEE transactions on information forensics and security , vol. 1, no. 2, pp. 125–143, 2006

    work page 2006

  5. [4]

    A. K. Jain and S. Z. Li, Handbook of face recognition. Springer, 2011, vol. 1

    work page 2011

  6. [5]

    Maltoni, D

    D. Maltoni, D. Maio, A. K. Jain, S. Prabhakar et al. , Handbook of fingerprint recognition. Springer, 2009, vol. 2

    work page 2009

  7. [6]

    A survey of palmprint recognition,

    A. Kong, D. Zhang, and M. Kamel, “A survey of palmprint recognition,” pattern recognition, vol. 42, no. 7, pp. 1408–1418, 2009

    2009

  8. [7]

    Bio- metric identification through hand geometry measurements,

    R. Sanchez-Reillo, C. Sanchez-Avila, and A. Gonzalez-Marcos, “Bio- metric identification through hand geometry measurements,” IEEE Transactions on pattern analysis and machine intelligence , vol. 22, no. 10, pp. 1168–1171, 2000

    work page 2000

Show all 64 references
  1. [8]

    Biometric recognition using 3D ear shape,

    P. Yan and K. W. Bowyer, “Biometric recognition using 3D ear shape,” IEEE Transactions on pattern analysis and machine intelligence, vol. 29, no. 8, pp. 1297–1308, 2007

    2007

  2. [9]

    Multimodal biometric system using face, ear and gait biometrics,

    A. P. Yazdanpanah, K. Faez, and R. Amirfattahi, “Multimodal biometric system using face, ear and gait biometrics,” in10th International Confer- ence on Information Science, Signal Processing and their Applications (ISSPA 2010). IEEE, 2010, pp. 251–254

    2010

  3. [10]

    Automatic on-line signature verification,

    V . S. Nalwa, “Automatic on-line signature verification,” Proceedings of the IEEE, vol. 85, no. 2, pp. 215–239, 1997

    1997

  4. [11]

    In- creasing the robustness of biometric templates for dynamic signature biometric systems,

    R. Tolosana, R. Vera-Rodriguez, J. Ortega-Garcia, and J. Fierrez, “In- creasing the robustness of biometric templates for dynamic signature biometric systems,” in 2015 International Carnahan Conference on Security Technology (ICCST). IEEE, 2015, pp. 229–234

    2015

  5. [12]

    Advances in anti-spoofing: from the perspective of asvspoof challenges,

    M. R. Kamble, H. B. Sailor, H. A. Patil, and H. Li, “Advances in anti-spoofing: from the perspective of asvspoof challenges,” APSIPA Transactions on Signal and Information Processing , vol. 9, p. e2, 2020

    2020

  6. [13]

    A review on state-of-the-art automatic speaker verification system from spoofing and anti-spoofing perspective,

    A. Chadha, A. Abdullah, L. Angeline, and S. Sivanesan, “A review on state-of-the-art automatic speaker verification system from spoofing and anti-spoofing perspective,” Indian Journal of Science and Technology , vol. 14, no. 40, pp. 3026–3050, 2021

    2021

  7. [14]

    High-quality text-to-speech synthesis: An overview,

    T. Dutoit, “High-quality text-to-speech synthesis: An overview,” Journal Of Electrical And Electronics Engineering Australia , vol. 17, no. 1, pp. 25–36, 1997

    1997

  8. [15]

    An overview of voice conversion systems,

    S. H. Mohammadi and A. Kain, “An overview of voice conversion systems,” Speech Communication, vol. 88, pp. 65–82, 2017

    2017

  9. [16]

    Asvspoof 2019: A large-scale public database of synthesized, converted and replayed speech,

    X. Wang, J. Yamagishi, M. Todisco, H. Delgado, A. Nautsch, N. Evans, M. Sahidullah, V . Vestman, T. Kinnunen, K. A. Lee et al. , “Asvspoof 2019: A large-scale public database of synthesized, converted and replayed speech,” Computer Speech & Language , vol. 64, p. 101114, 2020

    2019

  10. [17]

    Robust features fusion for text independent speaker verification enhancement in noisy environ- ments,

    M. Mohammadi and H. R. S. Mohammadi, “Robust features fusion for text independent speaker verification enhancement in noisy environ- ments,” in 2017 Iranian Conference on Electrical Engineering (ICEE) . IEEE, 2017, pp. 1863–1868

    2017

  11. [18]

    Automatic speaker verification systems and spoof detection techniques: review and analysis,

    A. Mittal and M. Dua, “Automatic speaker verification systems and spoof detection techniques: review and analysis,” International Journal of Speech Technology, pp. 1–30, 2022

    2022

  12. [19]

    Constant Q cepstral coeffi- cients: A spoofing countermeasure for automatic speaker verification,

    M. Todisco, H. Delgado, and N. Evans, “Constant Q cepstral coeffi- cients: A spoofing countermeasure for automatic speaker verification,” Computer Speech & Language , vol. 45, pp. 516–535, 2017

    2017

  13. [20]

    Tempo- ral variability and multi-viewed self-supervised representations to tackle the ASVspoof5 deepfake challenge,

    Y . Xie, X. Wang, Z. Wang, R. Fu, Z. Wen, H. Cheng, and L. Ye, “Tempo- ral variability and multi-viewed self-supervised representations to tackle the ASVspoof5 deepfake challenge,” arXiv preprint arXiv:2408.06922 , 2024

    2024 arXiv

  14. [21]

    V oice spoofing detection using a neural networks assembly considering spectrograms and mel frequency cepstral coefficients,

    C. A. Hern ´andez-Nava, E. A. Rinc ´on-Garc´ıa, P. Lara-Vel´azquez, S. G. de Los-Cobos-Silva, M. A. Guti ´errez-Andrade, and R. A. Mora- Guti´errez, “V oice spoofing detection using a neural networks assembly considering spectrograms and mel frequency cepstral coefficients,”Pe...

    2023

  15. [22]

    V oice spoofing detection for multiclass attack classification using deep learning,

    J. Boyd, M. Fahim, and O. Olukoya, “V oice spoofing detection for multiclass attack classification using deep learning,” Machine Learning With Applications, vol. 14, p. 100503, 2023

    2023

  16. [23]

    BUT systems and anal- yses for the ASVspoof 5 challenge,

    J. Rohdin, L. Zhang, O. Plchot, V . Stan ˇek, D. Mihola, J. Peng, T. Stafy- lakis, D. Beveraki, A. Silnova, J. Brukner et al., “BUT systems and anal- yses for the ASVspoof 5 challenge,” arXiv preprint arXiv:2408.11152 , 2024

    2024 arXiv

  17. [24]

    Stc antispoofing systems for the asvspoof2019 challenge,

    G. Lavrentyeva, S. Novoselov, A. Tseren, M. V olkova, A. Gorlanov, and A. Kozlov, “Stc antispoofing systems for the asvspoof2019 challenge,” arXiv preprint arXiv:1904.05576 , 2019

    1904 arXiv

  18. [25]

    Compact time-domain represen- tation for logical access spoofed audio,

    M. Karo, A. Yeredor, and I. Lapidot, “Compact time-domain represen- tation for logical access spoofed audio,” IEEE/ACM Transactions on Audio, Speech, and Language Processing , 2023

    2023

  19. [26]

    Wav2Vec 2.0: A framework for self-supervised learning of speech representations,

    A. Baevski, Y . Zhou, A. Mohamed, and M. Auli, “Wav2Vec 2.0: A framework for self-supervised learning of speech representations,” Advances in neural information processing systems, vol. 33, pp. 12 449– 12 460, 2020

    2020

  20. [27]

    WavLM model ensemble for audio deepfake detection,

    D. Combei, A. Stan, D. Oneata, and H. Cucu, “WavLM model ensemble for audio deepfake detection,” arXiv preprint arXiv:2408.07414 , 2024

    2024 arXiv

  21. [28]

    SZU-AFS antispoofing system for the ASVspoof 5 challenge,

    Y . Xu, J. Zhong, S. Zheng, Z. Liu, and B. Li, “SZU-AFS antispoofing system for the ASVspoof 5 challenge,”arXiv preprint arXiv:2408.09933, 2024

    2024 arXiv

  22. [29]

    One-class learning towards synthetic voice spoofing detection,

    Y . Zhang, F. Jiang, and Z. Duan, “One-class learning towards synthetic voice spoofing detection,” IEEE Signal Processing Letters , vol. 28, pp. 937–941, 2021

    2021

  23. [30]

    Additive margin softmax for face verification,

    F. Wang, J. Cheng, W. Liu, and H. Liu, “Additive margin softmax for face verification,” IEEE Signal Processing Letters , vol. 25, no. 7, pp. 926–930, 2018

    2018

  24. [31]

    Gender recognition from speech signal using CNN, KNN, SVM and RF,

    E. Y ¨ucesoy, “Gender recognition from speech signal using CNN, KNN, SVM and RF,” Procedia Computer Science , vol. 235, pp. 2251–2257, 2024

    2024

  25. [32]

    V oice-based gender recognition using neural network,

    K. Chachadi and S. Nirmala, “V oice-based gender recognition using neural network,” in Information and Communication Technology for Competitive Strategies (ICTCS 2020) ICT: Applications and Social Interfaces. Springer, 2022, pp. 741–749

    2020

  26. [33]

    V oice based gender classification using machine learning,

    A. Raahul, R. Sapthagiri, K. Pankaj, and V . Vijayarajan, “V oice based gender classification using machine learning,” in IOP Conference Series: Materials Science and Engineering , vol. 263, no. 4. IOP Publishing, 2017, p. 042083. 11

    2017

  27. [35]

    Effects of waveform pmf on anti-spoofing detection,

    I. Lapidot and J.-F. Bonastre, “Effects of waveform pmf on anti-spoofing detection,” in Interspeech 2019. ISCA, 2019, pp. 2853–2857

    2019

  28. [36]

    Meaningful embeddings for ex- plainable countermeasures,

    M. Karo, A. Yeredor, and I. Lapidot, “Meaningful embeddings for ex- plainable countermeasures,” in The Speaker and Language Recognition Workshop (Odyssey 2024), 2024, pp. 151–157

    2024

  29. [37]

    Speech database and protocol validation using waveform entropy

    I. Lapidot, H. Delgado, M. Todisco, N. W. Evans, and J.-F. Bonastre, “Speech database and protocol validation using waveform entropy.” in INTERSPEECH, 2018, pp. 2773–2777

    2018

  30. [38]

    The Quadratic-Chi Histogram Distance Family,

    O. Pele and M. Werman, “The Quadratic-Chi Histogram Distance Family,” in Computer Vision–ECCV 2010: 11th European Conference on Computer Vision, Heraklion, Crete, Greece, September 5-11, 2010, Proceedings, Part II 11 . Springer, 2010, pp. 749–762

    2010

  31. [39]

    Template matching using fast normalized cross correlation,

    K. Briechle and U. D. Hanebeck, “Template matching using fast normalized cross correlation,” in Optical pattern recognition XII , vol

  32. [40]

    Neue begr ¨undung der theorie quadratischer formen von unendlichvielen ver ¨anderlichen

    E. Hellinger, “Neue begr ¨undung der theorie quadratischer formen von unendlichvielen ver ¨anderlichen.” Journal f ¨ur die reine und angewandte Mathematik, vol. 1909, no. 136, pp. 210–271, 1909

    1909

  33. [41]

    XGBoost: extreme gradient boosting,

    T. Chen, “XGBoost: extreme gradient boosting,” R package version 0.4- 2, vol. 1, no. 4, 2015

    2015

  34. [42]

    Skip connections eliminate singularities,

    A. E. Orhan and X. Pitkow, “Skip connections eliminate singularities,” arXiv preprint arXiv:1701.09175 , 2017

    2017 arXiv

  35. [43]

    SpeechBrain: A general- purpose speech toolkit,

    M. Ravanelli, T. Parcollet, P. Plantinga, A. Rouhe, S. Cornell, L. Lu- gosch, C. Subakan, N. Dawalatabad, A. Heba, J. Zhong, J.-C. Chou, S.-L. Yeh, S.-W. Fu, C.-F. Liao, E. Rastorgueva, F. Grondin, W. Aris, H. Na, Y . Gao, R. D. Mori, and Y . Bengio, “SpeechBrain: A general- p...

    2021 arXiv

  36. [44]

    ECAPA-TDNN: emphasized channel attention, propagation and aggregation in TDNN based speaker verification,

    B. Desplanques, J. Thienpondt, and K. Demuynck, “ECAPA-TDNN: emphasized channel attention, propagation and aggregation in TDNN based speaker verification,” in Interspeech 2020, H. Meng, B. Xu, and T. F. Zheng, Eds. ISCA, 2020, pp. 3830–3834

    2020

  37. [45]

    ASVspoof 2019: Automatic speaker verification spoofing and coun- termeasures challenge evaluation plan,

    J. Yamagishi, M. Todisco, M. Sahidullah, H. Delgado, X. Wang, N. Evans, T. Kinnunen, K. A. Lee, V . Vestman, and A. Nautsch, “ASVspoof 2019: Automatic speaker verification spoofing and coun- termeasures challenge evaluation plan,” ASV Spoof, vol. 13, 2019

    2019

  38. [46]

    t-EER: Parameter-free tandem evaluation of countermeasures and biometric comparators,

    T. H. Kinnunen, K. A. Lee, H. Tak, N. Evans, and A. Nautsch, “t-EER: Parameter-free tandem evaluation of countermeasures and biometric comparators,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023

    2023

  39. [47]

    Out of a hundred trials, how many errors does your speaker verifier make?

    N. Br ¨ummer, L. Ferrer, and A. Swart, “Out of a hundred trials, how many errors does your speaker verifier make?” arXiv preprint arXiv:2104.00732, 2021

    2021 arXiv

  40. [48]

    Tandem assessment of spoofing countermeasures and automatic speaker verifi- cation: Fundamentals,

    T. Kinnunen, H. Delgado, N. Evans, K. A. Lee, V . Vestman, A. Nautsch, M. Todisco, X. Wang, M. Sahidullah, J. Yamagishi et al. , “Tandem assessment of spoofing countermeasures and automatic speaker verifi- cation: Fundamentals,” IEEE/ACM Transactions on Audio, Speech, and Lang...

    2020

  41. [49]

    a-DCF: an architecture agnostic metric with application to spoofing-robust speaker verification,

    H. jin Shim, J. weon Jung, T. Kinnunen, N. Evans, J.-F. Bonastre, and I. Lapidot, “a-DCF: an architecture agnostic metric with application to spoofing-robust speaker verification,” in The Speaker and Language Recognition Workshop (Odyssey 2024) , 2024, pp. 158–164

    2024

  42. [50]

    The NIST speaker recognition evaluation–overview, methodology, sys- tems, results, perspective,

    G. R. Doddington, M. A. Przybocki, A. F. Martin, and D. A. Reynolds, “The NIST speaker recognition evaluation–overview, methodology, sys- tems, results, perspective,” Speech communication, vol. 31, no. 2-3, pp. 225–254, 2000

    2000

  43. [51]

    t-DCF: a detection cost function for the tandem assessment of spoofing countermeasures and automatic speaker verification,

    T. Kinnunen, K. A. Lee, H. Delgado, N. Evans, M. Todisco, M. Sahidul- lah, J. Yamagishi, and D. A. Reynolds, “t-DCF: a detection cost function for the tandem assessment of spoofing countermeasures and automatic speaker verification,” arXiv preprint arXiv:1804.09618 , 2018

    2018 arXiv

  44. [52]

    ASVspoof 5 evaluation plan,

    H. Delgado, N. Evans, J.-w. Jung, T. Kinnunen, I. Kukanov, K. A. Lee, X. Liu, H.-j. Shim, M. Sahidullah, H. Tak et al., “ASVspoof 5 evaluation plan,” 2024

    2024

  45. [53]

    Model evaluation, model selection, and algorithm selection in machine learning,

    S. Raschka, “Model evaluation, model selection, and algorithm selection in machine learning,” arXiv preprint arXiv:1811.12808 , 2018

    2018 arXiv

  46. [54]

    Confidence intervals for evaluation in machine learning

    L. Ferrer and P. Riera, “Confidence intervals for evaluation in machine learning.” [Online]. Available: https://github.com/luferrer/ ConfidenceIntervals

  47. [55]

    Estimating the confidence interval of expected performance curve in biometric authentication using joint bootstrap,

    N. Poh and S. Bengio, “Estimating the confidence interval of expected performance curve in biometric authentication using joint bootstrap,” in 2007 IEEE International Conference on Acoustics, Speech and Signal Processing-ICASSP’07, vol. 2. IEEE, 2007, pp. II–137

    2007

  48. [56]

    Benchmarking non-parametric statistical tests,

    M. Keller, S. Bengio, and S. Wong, “Benchmarking non-parametric statistical tests,” Advances in neural information processing systems , vol. 18, 2005

    2005

  49. [57]

    Principal component analysis,

    H. Abdi and L. J. Williams, “Principal component analysis,” Wiley interdisciplinary reviews: computational statistics, vol. 2, no. 4, pp. 433– 459, 2010

    2010

  50. [58]

    UMAP: Uniform manifold approximation and projection for dimension reduction,

    L. McInnes, J. Healy, and J. Melville, “UMAP: Uniform manifold approximation and projection for dimension reduction,” arXiv preprint arXiv:1802.03426, 2018

    2018 arXiv

  51. [59]

    Borderline over-sampling for imbalanced data classification,

    H. M. Nguyen, E. W. Cooper, and K. Kamei, “Borderline over-sampling for imbalanced data classification,” International Journal of Knowledge Engineering and Soft Data Paradigms , vol. 3, no. 1, pp. 4–21, 2011

    2011

  52. [60]

    Towards theoretically understanding why SGD generalizes better than adam in deep learning,

    P. Zhou, J. Feng, C. Ma, C. Xiong, S. C. H. Hoi et al. , “Towards theoretically understanding why SGD generalizes better than adam in deep learning,” Advances in Neural Information Processing Systems , vol. 33, pp. 21 285–21 296, 2020

    2020

  53. [61]

    To what extent can asv systems naturally defend against spoofing attacks?

    J. weon Jung, X. Wang, N. Evans, S. Watanabe, H. jin Shim, H. Tak, S. Arora, J. Yamagishi, and J. S. Chung, “To what extent can asv systems naturally defend against spoofing attacks?” in Interspeech 2024, 2024, pp. 3240–3244

    2024

  54. [62]

    Automatic speaker verification spoofing and countermeasures chal- lenge,

    “Automatic speaker verification spoofing and countermeasures chal- lenge,” https://www.asvspoof.org/index2019.html, accessed: 2024-11- 11

    2024

  55. [63]

    Deep residual learning for image recognition,

    K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2016, pp. 770–778. Avishai Weizman Avishai Weizman received his B.Sc. in Electrical Engineering from Afeka Te...

    2016

  56. [4387]

    SPIE, 2001, pp. 95–102

    2001

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.