VoxCeleb: A Large-Scale Speaker Identification Dataset
Open accessThe DOI matches its Crossref record; 10 of the 10 references it cites were found in a registry or an index; none of the 16 sources its claims cite is checked for support yet.
Abstract
Most existing datasets for speaker identification contain samples obtained under quite constrained conditions, and are usually hand-annotated, hence limited in size. The goal of this paper is to generate a large scale text-independent speaker identification dataset collected 'in the wild'. We make two contributions. First, we propose a fully automated pipeline based on computer vision techniques to create the dataset from open-source media. Our pipeline involves obtaining videos from YouTube; performing active speaker verification using a two-stream synchronization Convolutional Neural Network (CNN), and confirming the identity of the speaker using CNN based facial recognition. We use this pipeline to curate VoxCeleb which contains hundreds of thousands of 'real world' utterances for over 1,000 celebrities. Our second contribution is to apply and compare various state of the art speaker identification techniques on our dataset to establish baseline performance. We show that a CNN based architecture obtains the best performance for both identification and verification.
Overview
CoChat Summary
TL;DR
The paper introduces VoxCeleb, a fully automated, large-scale speaker identification dataset built from YouTube videos “in the wild,” with 1,251 celebrities and over 100,000 utterances. It also benchmarks speaker recognition methods and reports that a CNN performs best on both identification and verification.
Key findings
- VoxCeleb contains over 100,000 utterances for 1,251 celebrities; the dataset is gender balanced, with 55% male speakers.
- The automated collection pipeline uses active speaker verification and face verification and was tuned to be highly precise: precision = 1.000 for both stages, with recall = 0.613 for active speaker verification and 0.726 for face verification.
- For speaker identification, the CNN achieves 80.5% top-1 and 92.1% top-5 accuracy, compared with 49.0% / 56.6% for I-vectors + SVM and 60.8% / 75.6% for I-vectors + PLDA + SVM.
- For speaker verification, the best reported result is CNN + Embedding with C min det = 0.71 and EER = 7.8%, better than I-vectors + PLDA (0.73, 8.8%) and GMM-UBM (0.80, 15.0%).
Methods
- Built VoxCeleb from YouTube videos using a fully automated multi-stage pipeline: candidate POI list, video download, face tracking, active speaker verification via SyncNet, and face verification via VGG Face CNN.
- Dataset scale/statistics reported: 1,251 POIs, 690 male POIs, 36 / 18 / 8 videos per POI (max/avg/min), 250 / 123 / 45 utterances per POI (max/avg/min), and 145.0 / 8.2 / 4.0 s utterance length (max/avg/min).
- Trained and evaluated a CNN on spectrograms with minimal preprocessing; identification used a 1,251-way classification setup, while verification compared I-vectors/PLDA, GMM-UBM, and CNN embedding approaches.
Limitations
- The paper states that the pipeline is designed to be conservative, and that this means “a lot of the downloaded videos” are discarded to reduce label errors.
- The paper notes that the collection starts from a list of 2,622 identities from VGG Face and uses YouTube search with “interview” appended, which is an evident source of selection bias toward publicly available, interview-like media.
- The dataset is described as coming from open-source media / YouTube, so it is limited to speakers and contexts available there; this is evident from the text, though not stated as a formal limitation.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 Valid · 1 Not valid · 12 Not checked · 37 Own results and interpretations
In the paper
“| Face verification | [30] | 1.000 | 0.726 |”
What the sources say
- Not checked yetcites [30] [30] O. M. Parkhi, A. Vedaldi, and A. Zisserman, 'Deep face recognition,' in Proceedings of the British Machine Vision Conference , 2015. -
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Unlike the SITW dataset, both audio and video for each speaker is released.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“| Active speaker verification | [31] | 1.000 | 0.613 |”
What the sources say
- Not checked yetcites [31] [31] P. Chakravarty and T. Tuytelaars, 'Cross-modal supervision for learning active speaker detection in video,' arXiv preprint arXiv:1603.08907 , 2016. -
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“reduces the number of parameters from 319M in VGG-M to 67M in our network”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“| # of videos per POI | 36 / 18 / 8 |”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
15 more claims
Create a free account to see all 50 claims and to verify the 16 references still open.
- Probably not a substantial claim. 9
- It may state two separate claims as one. 23
- The quote only partly states this claim. 5
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“Meanwhile, the emergence of large-scale VSA datasets [28, 29, 30, 31, 32] promoted the further development of deep learning based VSA research.”
From Deep learning for visual speech analysis: A survey (2024) · cites this paper as [31]
In the citing paper
“Although it is mainly built for speaker identification, it also can be used for VSG.”
From Deep learning for visual speech analysis: A survey (2024) · cites this paper as [31]
In the citing paper
“It contains over 100k utterances from 1,251 celebrities.”
From Deep learning for visual speech analysis: A survey (2024) · cites this paper as [31]
In the citing paper
“VoxCeleb1 [31] is a large-scale text-independent audio-visual dataset collected from open-source YouTube media.”
From Deep learning for visual speech analysis: A survey (2024) · cites this paper as [31]
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S., & Szegedy, C. (2015) · arXiv · doi:10.48550/arxiv.1502.03167 · Cited by 27,818 · cited 1×
“The network is trained using batch normalisation [40] and all hyper-parameters (e.”
MethodBatch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S. (2024) · arXiv (Cornell University) · doi:10.57702/o9raffed · Cited by 14,214
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1512.03385 · Cited by 4,196
One millisecond face alignment with an ensemble of regression trees
Kazemi, V., & Sullivan, J. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.241 · Cited by 2,831 · cited 1×
“Facial landmark positions are detected for each face detection using the regression tree based method of [33].”
MethodCNN architectures for large-scale audio classification
Hershey, S., Chaudhuri, S., Ellis, D. P. W., Gemmeke, J. F., Jansen, A., Moore, R. C., Plakal, M., Platt, D., Saurous, R. A., Seybold, B., et al. (2017) · 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2017.7952132 · Cited by 2,023
MatConvNet - Convolutional Neural Networks for MATLAB
Vedaldi, A., & Lenc, K. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1412.4564 · Cited by 578 · cited 1×
“Our implementation is based on the deep learning toolbox MatConvNet [39] and trained on a NVIDIA TITAN X GPU.”
MethodCNN Architectures for Large-Scale Audio ClassificationPreprint
Hershey, S., Chaudhuri, S., Ellis, D. P. W., Gemmeke, J. F., Jansen, A., Moore, R. C., Plakal, M., Platt, D., Saurous, R. A., Seybold, B., et al. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1609.09430 · Cited by 217 · cited 1×
“This has led to a very recent shift from hand-crafted features to the domain of deep CNNs which can be applied to higher dimensional inputs [13, 14] and for speaker iden-tification [15].”
BackgroundCross-Modal Supervision for Learning Active Speaker Detection in Video
Chakravarty, P., & Tuytelaars, T. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-46454-1_18 · Cited by 70 · cited 2×
“Precision-recall curves for both tasks on their respective benchmark datasets [30, 31] are shown in Figure 1 right, and the values at the operating point are given in Table 3.”
BackgroundCross-modal Supervision for Learning Active Speaker Detection in VideoPreprint
Chakravarty, P., & Tuytelaars, T. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1603.08907 · Cited by 1 · cited 2×
“Precision-recall curves for both tasks on their respective benchmark datasets [30, 31] are shown in Figure 1 right, and the values at the operating point are given in Table 3.”
Background
Research works printed but not linked: 28
- [2][2] A. Krizhevsky, I. Sutskever, and G. E. Hinton, 'ImageNet classification with deep convolutional neural networks,' in Advances in Neural Information Processing Systems , pp. 1106-1114, 2012. -
- [5][5] M. McLaren, L. Ferrer, D. Castan, and A. Lawson, 'The speakers in the wild (SITW) speaker recognition database,' INTERSPEECH , vol. 2016, 2016. -
- [6][6] O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, S. Huang, A. Karpathy, A. Khosla, M. Bernstein, A. Berg, and F. Li, 'Imagenet large scale visual recognition challenge,' International Journal of Computer Vision , 2015. -
- [7][7] D. A. Reynolds, T. F. Quatieri, and R. B. Dunn, 'Speaker verification using adapted gaussian mixture models,' Digital signal processing , vol. 10, no. 1-3, pp. 19-41, 2000. -
- [8][8] D. A. Reynolds and R. C. Rose, 'Robust text-independent speaker identification using gaussian mixture speaker models,' IEEE transactions on speech and audio processing , vol. 3, no. 1, pp. 7283, 1995. -
- [10][10] N. Dehak, P. J. Kenny, R. Dehak, P. Dumouchel, and P. Ouellet, 'Front-end factor analysis for speaker verification,' IEEE Transactions on Audio, Speech, and Language Processing , vol. 19, no. 4, pp. 788-798, 2011. -
- [11][11] U. H. Yapanel, X. Zhang, and J. H. Hansen, 'High performance digit recognition in real car environments.,' in INTERSPEECH , 2002. -
- [12][12] J. H. Hansen, R. Sarikaya, U. H. Yapanel, and B. L. Pellom, 'Robust speech recognition in noise: an evaluation using the spine corpus.,' in INTERSPEECH , pp. 905-908, 2001. -
- [13][13] T. N. Sainath, R. J. Weiss, A. W. Senior, K. W. Wilson, and O. Vinyals, 'Learning the speech front-end with raw waveform CLDNNs,' in INTERSPEECH , pp. 1-5, 2015. -
- [15][15] Y. Lukic, C. Vogt, O. D¨ urr, and T. Stadelmann, 'Speaker identification and clustering using convolutional neural networks,' in IEEE 26th International Workshop on Machine Learning for Signal Processing (MLSP) , pp. 1-6, IEEE, 2016. -
- [16][16] D. van der Vloed, J. Bouten, and D. A. van Leeuwen, 'NFIFRITS: a forensic speaker recognition database and some first experiments,' in The Speaker and Language Recognition Workshop , 2014. -
- [17][17] J. Hennebert, H. Melin, D. Petrovska, and D. Genoud, 'POLYCOST: a telephone-speech database for speaker recognition,' Speech communication , vol. 31, no. 2, pp. 265-270, 2000. -
- [18][18] J. B. Millar, J. P. Vonwiller, J. M. Harrington, and P. J. Dermody, 'The Australian national database of spoken language,' in Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing , vol. 1, pp. I-97, IEEE, 1994. -
- [19][19] J. S. Garofolo, L. F. Lamel, W. M. Fisher, J. G. Fiscus, and D. S. Pallett, 'DARPA TIMIT acoustic-phonetic continous speech corpus CD-ROM. NIST speech disc 1-1.1,' NASA STI/Recon technical report , vol. 93, 1993. -
- [21][21] R. Woo, A. Park, and T. J. Hazen, 'The MIT Mobile Device Speaker Verification Corpus: Data collection and preliminary experiments,' The Speaker and Language Recognition Workshop , 2006. -
- [23][23] A. Janin, D. Baron, J. Edwards, D. Ellis, D. Gelbart, N. Morgan, B. Peskin, T. Pfau, E. Shriberg, A. Stolcke, et al. , 'The ICSI meeting corpus,' in Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing , vol. 1, IEEE, 2003. -
- [24][24] I. McCowan, J. Carletta, W. Kraaij, S. Ashby, S. Bourban, M. Flynn, M. Guillemot, T. Hain, J. Kadlec, V. Karaiskos, et al. , 'The AMI meeting corpus,' in International Conference on Methods and Techniques in Behavioral Research , vol. 88, 2005. -
- [25][25] P. Bell, M. J. Gales, T. Hain, J. Kilgour, P. Lanchantin, X. Liu, A. McParland, S. Renals, O. Saz, M. Wester, et al. , 'The MGB challenge: Evaluating multi-genre broadcast media recognition,' in IEEE Workshop on Automatic Speech Recognition and Understanding , pp. 687-693, IEEE, 2015. -
- [27][27] J. J. Godfrey, E. C. Holliman, and J. McDaniel, 'Switchboard: Telephone speech corpus for research and development,' in Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing , vol. 1, pp. 517-520, IEEE, 1992. -
- [28][28] W. M. Fisher, G. R. Doddington, and K. M. Goudie-Marshall, 'The DARPA speech recognition research database: specifications and status,' in Proc. DARPA Workshop on speech recognition , pp. 93-99, 1986. -
- [30][30] O. M. Parkhi, A. Vedaldi, and A. Zisserman, 'Deep face recognition,' in Proceedings of the British Machine Vision Conference , 2015. -
- [32][32] D. E. King, 'Dlib-ml: A machine learning toolkit,' The Journal of Machine Learning Research , vol. 10, pp. 1755-1758, 2009. -
- [34][34] J. S. Chung and A. Zisserman, 'Lip reading in the wild,' in Proceedings of the Asian Conference on Computer Vision , 2016. -
- [35][35] M. Everingham, J. Sivic, and A. Zisserman, 'Taking the bite out of automatic naming of characters in TV video,' Image and Vision Computing , vol. 27, no. 5, 2009. -
- [36][36] J. S. Chung and A. Zisserman, 'Out of time: automated lip sync in the wild,' in Workshop on Multi-view Lip-reading, ACCV , 2016. -
- [37][37] K. Chatfield, K. Simonyan, A. Vedaldi, and A. Zisserman, 'Return of the devil in the details: Delving deep into convolutional nets,' in Proceedings of the British Machine Vision Conference , 2014. -
- [38][38] S. Chopra, R. Hadsell, and Y. LeCun, 'Learning a similarity metric discriminatively, with application to face verification,' in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , vol. 1, pp. 539-546, IEEE, 2005. -
- [41][41] S. Ioffe, 'Probabilistic linear discriminant analysis,' in Proceedings of the European Conference on Computer Vision , pp. 531542, Springer, 2006.
Cited by
Items 901–1,000 of 1,008
Disentangled Speaker Embedding for Robust Speaker Verification
Yi, L., & Mak, M.-W. (2022) · IEEE International Conference on Acoustics, Speech, and Signal Processing · doi:10.1109/icassp43922.2022.9747778 · Cited by 8
“A “ × ” in the fourth column means that we used the development sets of VoxCeleb1&2 and their augmented sets to train the speaker embedding networks G spk without domain adaptation.”
BackgroundA tibetan-dependent speaker recognition method based on deep learning
Gan, Z., Yu, Y., & Luo, M. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-12540-9 · Cited by 8
“Speaker recognition can be divided into speaker identification (SI) and speaker verification (SV) in application [17].”
BackgroundWang, F., Song, Z., Jiang, H., & Xu, B. (2021) · IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU) · doi:10.1109/asru51503.2021.9687918 · Cited by 9
Complex-Valued Time-Frequency Self-Attention for Speech Dereverberation
Kothapally, V., & Hansen, J. H. L. (2022) · Interspeech (USB) · doi:10.21437/interspeech.2022-11277 · Cited by 13
“In order to evaluate the performance improvements for speaker verification (SV), we train a conventional X-vector extractor [32] on the Voxceleb 1+2 datasets [33, 34].”
MethodMemory Storable Network Based Feature Aggregation for Speaker Representation Learning
Gu, B., Guo, W., & Zhang, J. (2023) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2022.3231709 · Cited by 15
“TheVoxCeleb1 dataset [45] includes short-duration speech collected from online videos using an automated pipeline under unconstrained conditions.”
MethodCross-modal dynamic sentiment annotation for speech sentiment analysis
Chen, J., Sun, C., Zhang, S., & Zeng, J. (2023) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2023.108598 · Cited by 10
Margin-Mixup: A Method for Robust Speaker Verification In Multi-Speaker Audio
Thienpondt, J., Madhu, N., & Demuynck, K. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10095305 · Cited by 12
Self-Supervised Audio-Visual Speaker Representation with Co-Meta Learning
Chen, H., Zhang, H., Wang, L., Lee, K. A., Liu, M., & Dang, J. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10096925 · Cited by 12
“(3); For evalution, the experiments are conducted on the Original, Extended and Hard Voxceleb1 test sets [21, 22].”
MethodAlam, J., Kang, W. H., & Fathan, A. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10096040 · Cited by 8
“In order to evaluate the speaker verification performance of our proposed system, we have conducted a set of experiments on the Voxceleb [17, 18] dataset.”
MethodSalim, S., Shahnawazuddin, S., & Ahmad, W. (2023) · Applied Acoustics · doi:10.1016/j.apacoust.2023.109412 · Cited by 8
An analytic study on clustering driven self-supervised speaker verification
Fathan, A., & Alam, J. (2024) · Pattern Recognition Letters · doi:10.1016/j.patrec.2024.01.024 · Cited by 8
A Corrective Learning Approach for Text-Independent Speaker Verification
Wen, Y., Zhou, T., Singh, R., & Raj, B. (2018) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2018.8461340 · Cited by 8
“The most effective method analyzes Mel-frequency spectrographic representations of recordings with a convolutional neural network (CNN) to derive their parametrizations [11, 12, 13, 14].”
MethodA Bayesian Attention Neural Network Layer for Speaker Recognition
Zhu, W., & Pelecanos, J. W. (2019) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2019.8682953 · Cited by 8
“The second portion of training data is a combination of NIST data and data extracted from the OpenSLR corpus [20] and the VoxCeleb Corpus [21].”
BackgroundMethodA Review on Face Reenactment Techniques
Dhere, S., Rathod, S. B., Aarankalle, S., Lad, Y., & Gandhi, M. (2020) · International Conference on Industry 4.0 Technology (I4Tech) · doi:10.1109/i4tech48345.2020.9102668 · Cited by 10
“Zakharov et al. [27] presented a GAN-based [28] system which performed meta-learning on the VoxCeleb [8, 9] dataset first and then was able to create face models of unseen people from just a few images.”
BackgroundAdversarial Transformation of Spoofing Attacks for Voice Biometrics
Gomez-Alanis, A., González, J. A., & Peinado, A. M. (2021) · IberSPEECH · doi:10.21437/iberspeech.2021-54 · Cited by 14
Large-scale singer recognition using deep metric learning: an experimental study
Hu, S., Liang, B., Chen, Z., Lu, X. Q., Zhao, E., & Lui, S. (2021) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn52387.2021.9533911 · Cited by 8
“Existing deep embedding models address speaker recognition on spoken voices as a classification task [4], [14], or using metric learning [2], [3], [5], [6].”
BackgroundMulti-Task Voice Activated Framework Using Self-Supervised Learning
Hussain, S., Nguyen, V. H., Zhang, S., & Visser, E. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9746409 · Cited by 12
Spoofing-Aware Speaker Verification with Unsupervised Domain Adaptation
Liu, X., Sahidullah, M., & Kinnunen, T. (2022) · The Speaker and Language Recognition Workshop (Odyssey) · doi:10.21437/odyssey.2022-12 · Cited by 8
Zaiem, S., Parcollet, T., & Essid, S. (2022) · Interspeech (USB) · doi:10.21437/interspeech.2022-10191 · Cited by 10
Autonomous Learning of Speaker Identity and WiFi Geofence From Noisy Sensor Data
Lu, C. X., Xiangli, Y., Zhao, P., Chen, C., Trigoni, N., & Markham, A. (2019) · IEEE Internet of Things Journal · doi:10.1109/jiot.2019.2926645 · Cited by 8
“In our experiment, we used the X-vector feature extractor trained on the augmented VoxCeleb corpus [35] (augmented with MUSAN [36]), and the PLDA backend initially trained on purely VoxCeleb.”
MethodAn end-to-end approach for the verification problem: learning the right distancePreprint
Monteiro, J., Albuquerque, I., Alam, J., Hjelm, R. D., & Falk, T. H. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2002.09469 · Cited by 8
“To this end, we make use of the recently introduced VoxCeleb corpus (Nagrani et al., 2017; Chung et al., 2018), corresponding to audio recordings of interviews taken from youtube videos, which means there’s no control over the acoustic conditions present in the data.”
MethodA Study of Few-Shot Audio ClassificationPreprint
Wolters, P., Careaga, C., Hutchinson, B., & Phillips, L. A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.01573 · Cited by 15
“Performing exhaustive experiments on VoxCeleb [14] and VCTK datasets, they obtain 5-way speaker classification accuracies of 91.5% and 96.5%, respectively.”
InfluentialSignificance of Speaker Embeddings and Temporal Context for Depression DetectionPreprint
Dumpala, S. H., Rodriguez, S., Rempel, S., Uher, R., & Oore, S. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2107.13969 · Cited by 14
“In this work, the GE2E network is pre-trained on the task of speaker verification by consolidating 3 different datasets i.e., LibriSpeech [37], VoxCeleb1 and VoxCeleb2 [38] with 1166 speakers, 1211 speakers and 5994 speakers, respectively.”
A Military Audio Dataset for Situational Awareness and Surveillance
Kim, J.-W., Yoon, C., & Jung, H. (2024) · Scientific Data · doi:10.1038/s41597-024-03511-w · Cited by 14
“However, recent audio and speech, as well as video datasets 13,14,29,37–40 derived from YouTube are generally deployed with .mp4 or .wav as well as .avi format, and various researchers mainly have used these datasets.”
BackgroundMethodA Data Perspective on Ethical Challenges in Voice Biometrics Research
Leschanowsky, A., Rusti, C., Quinlan, C., Pnacekova, M., Gorce, L., & Hutiri, W. T. (2024) · IEEE Transactions on Biometrics Behavior and Identity Science · doi:10.1109/tbiom.2024.3446846 · Cited by 13
Descriptor: Voice Pre-Processing and Quality Assessment Dataset (VPQAD)
Ahmed, A., Khondkar, M. J. A., Herrick, A., Schuckers, S., & Imtiaz, M. H. (2024) · IEEE data descriptions. · doi:10.1109/ieeedata.2024.3493798 · Cited by 10
“While VoxCeleb is particularly useful for speaker recognition and verification tasks, its focus on speaker identification limits its broader applicability to general ASR research [14].”
Mamba-based Segmentation Model for Speaker Diarization
Plaquet, A., Tawara, N., Delcroix, M., Horiguchi, S., Ando, A., & Araki, S. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10889446 · Cited by 7
“2) Embedding extraction and clustering: For the second part of the pipeline, we use the process described in [3] with a ResNet [20] model pre-trained on VoxCeleb [21]–[23] for feature extraction.”
Sequence-to-Sequence Neural Diarization With Automatic Speaker Detection and Representation
Cheng, M., Lin, Y., & Li, M. (2025) · IEEE Transactions on Audio Speech and Language Processing · doi:10.1109/taslpro.2025.3581032 · Cited by 16
“Accordingly, they obtain 1.17%, 0.81%, and 0.34% equal error rates (EERs) on the Vox-O [79] trial.”
InfluentialAniFaceDiff: Animating stylized avatars via parametric conditioned diffusion models
Chen, K., Seneviratne, S., Wang, W., Hu, D., Saha, S., Hasan, M. T., Rasnayaka, S., Malepathirana, T., Gong, M., & Halgamuge, S. K. (2025) · Pattern Recognition · doi:10.1016/j.patcog.2025.112017 · Cited by 13
“We utilized the subset of VoxCeleb dataset [59] after preprocessing as described in [5] for training and test.”
Dynamic Stereotype Theory Induced Micro-expression Recognition with Oriented Deformation
Zhang, B., Wang, X., Wang, C., & He, G. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.01000 · Cited by 12
SIBERIA: A Self-Sovereign Identity and Multi-Factor Authentication Framework for Industrial Access
Paredes-García, D., Fernández-Carrasco, J. Á., López, J. A. M., Vásquez-Correa, J. C., Yoldi, I. J., Moreno-Acevedo, S. A., González-Docasal, A., Arzelus, H., Álvarez, A., & Loinaz, Y. D. D. (2025) · Applied Sciences · doi:10.3390/app15158589 · Cited by 10
Image Animation with Perturbed Masks
Shalev, Y., & Wolf, L. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.00363 · Cited by 11
“VoxCeleb is an audio-visual dataset consist of short videos of talking faces, introduced by [14].”
BackgroundAge Estimation in Short Speech Utterances Based on Bidirectional Gated-Recurrent Neural Networks
Badr, A. A., & Hassan, A. K. A. (2021) · Engineering and Technology Journal · doi:10.30684/etj.v39i1b.1905 · Cited by 7
“The dataset includes diverse background noise and varied utterance duration because it comprises various utterances of celebrities from YouTube [25].”
BackgroundCoherent Image Animation Using Spatial-Temporal Correspondence
Wang, W., Zhou, W., Bao, J., & Li, H. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3160297 · Cited by 9
“methods, we train and test our network on four datasets, the Tai-Chi-HD dataset [1], the VoxCeleb dataset [43], the BAIR dataset [44] and UCF-101 [45], with various objects, such as human postures, faces and robot arms.”
MethodSpeech-based gender recognition using linear prediction and mel-frequency cepstral coefficients
Ali, Y. M., Noorsal, E., Mokhtar, N. F., Saad, S. Z. M., Abdullah, M. H., & Chin, L. C. (2022) · Indonesian Journal of Electrical Engineering and Computer Science · doi:10.11591/ijeecs.v28.i2.pp753-761 · Cited by 7
“Tang et al. [13] explored on gender and nationality information in public database called VoxCeleb1 [14] to improve speaker recognition using a mining algorithm based on those two embedding features proposed using CNN classification frameworks achieved more than 98% accuracy of gender recognition on different spectrog…”
Learned Source and Channel Coding for Talking-Head Semantic Transmission
Yue, W., Dai, J., Wang, S., Si, Z., & Niu, K. (2023) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc55385.2023.10118851 · Cited by 8
“We evaluate the results over the AWGN channel at channel SNR = 10dB, LPIPS is close to 0.07 and 0.08 for VoxCeleb1 and VoxCeleb2.”
MethodInfluentialAugmentation Techniques for Adult-Speech to Generate Child-Like Speech Data Samples at Scale
Yiwere, M. Y., Barcovschi, A., Jain, R., Cucu, H., & Corcoran, P. (2023) · IEEE Access · doi:10.1109/access.2023.3317360 · Cited by 9
“and well-annotated datasets [31], [32], [33], [34], [35], [36].”
Constant Q Cepstral Coefficients for Automatic Speaker Verification System for Dysarthria Patients
Salim, S., & Ahmad, W. (2023) · Circuits Systems and Signal Processing · doi:10.1007/s00034-023-02505-0 · Cited by 7
“The Voxceleb1 database [8] is utilized for training the speaker models during the training phase.”
Chen, Y.-L., Zeng, R.-L., & Xiong, S.-W. (2024) · IEEE transactions on consumer electronics · doi:10.1109/tce.2023.3323684 · Cited by 7
“We apply VoxCeleb [71] test set and VoxCeleb2 [72] to evaluate our method, while the test faces are not seen at the training time.”
MethodInfluentialA multimodal screening system for elderly neurological diseases based on deep learning
Park, S., No, C., Kim, S., Han, K., Jung, J., Kwon, K., & Lee, M. (2023) · Scientific Reports · doi:10.1038/s41598-023-48071-y · Cited by 7
Protecting Your Voice from Speech Synthesis Attacks
Liu, Z., Zhang, Y., & Miao, C. (2023) · Annual Computer Security Applications Conference · doi:10.1145/3627106.3627183 · Cited by 19
“The AutoVC model used in this paper is trained on the VoxCeleb1 [46] and LibriSpeech datasets [49].”
MethodFairness and Privacy in Voice Biometrics: A Study of Gender Influences Using wav2vec 2.0
Chouchane, O., Panariello, M., Galdi, C., Todisco, M., & Evans, N. (2023) · International Conference of the Biometrics Special Interest Group (BIOSIG) · doi:10.1109/biosig58226.2023.10345975 · Cited by 7
“We used the VoxCeleb1 and VoxCeleb2 speaker recognition databases [4], [14].”
MethodAnti-Compression Contrastive Facial Forgery Detection
Huang, J., Du, C., Zhu, X., Ma, S., Nepal, ., & Xu, C. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3347103 · Cited by 7
“The dataset contains 10,000 real videos collected from VoxCeleb [36] and generates 10,000 animation videos, each featuring ten specific actions such as blinking and nodding (1,000 videos for each action).”
BackgroundMulti speaker text-to-speech synthesis using generalized end-to-end loss function
Nazir, O., Malik, A., Singh, S., & Pathan, A. K. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-18121-2 · Cited by 8
AudioRepInceptionNeXt: A lightweight single-stream architecture for efficient audio recognition
Lau, K. W., Rehman, Y. A. U., & Po, L.-M. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.127432 · Cited by 8
“…revolutionary impact on numerous audio understanding domains, including but not limited to predictive tasks like sound event classification [24], the direction of voice prediction [25], speech command recognition [26], speaker identification [27], and generative tasks such as music generation [28].”
BackgroundReidentification of Participants in Shared Clinical Data Sets: Experimental Study
Wiepert, D. A., Malin, B., Duffy, J. R., Utianski, R. L., Stricker, J. L., Jones, D. T., & Botha, H. (2024) · JMIR AI · doi:10.2196/52054 · Cited by 11
“The VoxCeleb data set has no privacy protections or additional consent processes in place given its public nature—all recordings come from interviews of celebrities posted on YouTube [23,35,36].”
BackgroundGe, Z., Xu, X., Guo, H., Wang, T., & Yang, Z. (2024) · Applied Acoustics · doi:10.1016/j.apacoust.2024.109929 · Cited by 8
Benchmark Dataset Dynamics, Bias and Privacy Challenges in Voice Biometrics Research
Rusti, C., Leschanowsky, A., Quinlan, C., Pňačeková, M., Gorce, L., & Toussaint, W. (2023) · IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb57857.2023.10449225 · Cited by 13
“The VoxCeleb datasets were constructed with a fully automated data processing pipeline from audio-visual media scraped from YouTube [27, 26].”
MethodPrototype Division for Self-Supervised Speaker Verification
Zhao, Z., Li, Z., Zhang, X., Wang, W., & Zhang, P. (2024) · IEEE Signal Processing Letters · doi:10.1109/lsp.2024.3377593 · Cited by 11
Profile-Error-Tolerant Target-Speaker Voice Activity Detection
Wang, D., Xiao, X., Kanda, N., Yousefi, M., Yoshioka, T., & Wu, J. (2024) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp48485.2024.10446475 · Cited by 7
“In the simulation process, we randomly mixed the utterances of 1-10 speakers from VoxCeleb1 [30] and VoxCeleb2 [31] with overlap ratios up to 30%.”
MethodDiarist: Streaming Speech Translation with Speaker Diarization
Yang, M., Kanda, N., Wang, X., Chen, J., Wang, P., Xue, J., Li, J., & Yoshioka, T. (2024) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp48485.2024.10446050 · Cited by 10
“After training the streaming ST model, we trained the t-vector model using VoxCeleb [37, 38], following the same training configuration used in [22].”
ChildAugment: Data augmentation methods for zero-resource children's speaker verification
Singh, V. P., Sahidullah, M., & Kinnunen, T. (2024) · The Journal of the Acoustical Society of America · doi:10.1121/10.0025178 · Cited by 18
A lightweight approach to real-time speaker diarization: from audio toward audio-visual data streams
Kynych, F., Červa, P., Žďánský, J., Svendsen, T. K., & Salvi, G. (2024) · EURASIP Journal on Audio Speech and Music Processing · doi:10.1186/s13636-024-00382-2 · Cited by 11
SQ-Whisper: Speaker-Querying Based Whisper Model for Target-Speaker ASR
Guo, P., Chang, X., Lv, H., Watanabe, S., & Xie, L. (2024) · IEEE Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2024.3513835 · Cited by 18
“Note that the output labels corresponding to the speaker prompts are excluded from the computation of the CE loss.”
Audio–visual self-supervised representation learning: A survey
Alsuwat, M. S., Al-Shareef, S., & Alghamdi, M. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.129750 · Cited by 10
LEAT: Towards robust deepfake disruption in real-world scenarios via Latent Ensemble Attack
Shim, J., & Yoon, H. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.127417 · Cited by 10
Eliminating Data Collection Bottleneck for Wake Word Engine Training Using Found and Synthetic Data
Ramanan, B., Drabeck, L. M., Woo, T. L., Cauble, T., & Rana, A. (2019) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata47090.2019.9006601 · Cited by 7
“Training with VoxCeleb1 [22] and VoxCeleb2 [23] datasets, which contain a cumulative of 7205 unique speakers with approximately 150 utterances per speaker, will most likely contribute towards a far-better performing Synthetic data-based WWE.”
BackgroundSegment Aggregation for Short Utterances Speaker Verification Using Raw Waveforms
Kim, S.-B., Jung, J.-W., Shim, H.-J., Kim, J.-H., & Yu, H.-J. (2020) · Interspeech (USB) · doi:10.21437/interspeech.2020-1564 · Cited by 9
Action Unit Driven Facial Expression Synthesis from a Single Image with Patch Attentive GAN
Zhao, Y. H., Yang, L., Pei, E., Oveneke, M. C., Alioscha‐Perez, M., Li, L., Jiang, D., & Sahli, H. (2021) · Computer Graphics Forum · doi:10.1111/cgf.14202 · Cited by 15
An Investigation of the Effectiveness of Phase for Audio Classification
Hidaka, S., Wakamiya, K., & Kaburagi, T. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9746037 · Cited by 8
Voxstructor: Voice Reconstruction from Voiceprint
Lu, P., Li, Q., Zhu, H., Sovernigo, G., & Lin, X. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-91356-4_20 · Cited by 9
Anchor voiceprint recognition in live streaming via RawNet-SA and gated recurrent unit
Yao, J., Zhang, J., Li, J., & Li, Z. (2021) · EURASIP Journal on Audio Speech and Music Processing · doi:10.1186/s13636-021-00234-3 · Cited by 7
“[12] directly extracted the voiceprint features using CNN”
MethodSpeech synthesis with face embeddings
Wu, X., Ji, S., Wang, J., & Guo, Y. (2022) · Applied Intelligence · doi:10.1007/s10489-022-03227-7 · Cited by 8
“01, CommonVoice [47], VoxCeleb1 [24] and VoxCeleb2 [46] datasets were added.”
BackgroundMethodInfluentialLearning Temporal Point Processes for Efficient Retrieval of Continuous Time Event Sequences
Gupta, V., Bedathur, S., & De, A. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i4.20317 · Cited by 22
Target Speaker Extraction by Fusing Voiceprint Features
Cheng, S., Shen, Y., & Wang, D. (2022) · Applied Sciences · doi:10.3390/app12168152 · Cited by 8
Abdulqader, A. H., Al-Haddad, S. A. R., Abdo, S., Abdulghani, A., & Natarajan, S. (2022) · International Conference on Emerging Smart Technologies and Applications (eSmarTA) · doi:10.1109/esmarta56775.2022.9935422 · Cited by 7
Few-shot re-identification of the speaker by social robots
Foggia, P., Greco, A., Roberto, A., Saggese, A., & Vento, M. (2022) · Autonomous Robots · doi:10.1007/s10514-022-10073-6 · Cited by 7
“In this section we report the results of the proposed speaker re-identification algorithm over VoxCeleb1.”
MethodInfluentialLi, Y., Zhang, X., & Sun, M. (2023) · Information Fusion · doi:10.1016/j.inffus.2023.02.030 · Cited by 8
Speaker Verification Based on Single Channel Speech Separation
Jin, R., Ablimit, M., & Hamdulla, A. (2023) · IEEE Access · doi:10.1109/access.2023.3287868 · Cited by 10
“Voxceleb1 [34] is a large-scale open source audio and video dataset released by the University of Oxford in 2017, only the audio data is used in this research.”
BackgroundIntegrating Emotion Recognition with Speech Recognition and Speaker Diarisation for Conversations
Wu, W., Zhang, C., & Woodland, P. C. (2023) · Interspeech (USB) · doi:10.21437/interspeech.2023-293 · Cited by 8
“The reference speaker embedding model was pre-trained on Voxceleb 1.0 [28].”
MethodInfluentialAn Automatic Speaker Clustering Pipeline for the Air Traffic Communication Domain
Khalil, D., Prasad, A., Motlíček, P., Zuluaga-Gómez, J., Nigmatulina, I., Madikeri, S. R., & Schuepbach, C. (2023) · Aerospace · doi:10.3390/aerospace10100876 · Cited by 7
“VoxCeleb 1 [35]: This is a dataset comprising 100,000 utterances by 1251 celebrities, which were extracted from videos uploaded to YouTube.”
BackgroundMicPro: Microphone-based Voice Privacy Protection
Xiao, S., Ji, X., Yan, C., Zheng, Z., & Xu, W. (2023) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3576915.3616616 · Cited by 13
“Our training dataset is Vox-Celeb1 [36] and LibriSpeech [38] (subset dev and train-clean-360 ), two English speaker identification datasets.”
MethodSemantic Face Compression for Metaverse: A Compact 3D Descriptor Based Approach
Li, B., Chen, B., Wang, Z., Wang, S., & Ye, Y. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3391612 · Cited by 8
“To train the descriptors compression network, the VoxCeleb dataset [24] is introduced as Authorized licensed use limited to the terms of the applicable license agreement with IEEE.”
MethodKinnunen, T., Hautamäki, R. G., Vestman, V., & Sahidullah, M. (2019) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2019.8683811 · Cited by 10
Unsupervised Learning of Landmarks by Descriptor Vector Exchange
Thewlis, J., Albanie, S., Bilen, H., & Vedaldi, A. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00646 · Cited by 69
“∗ report a more conservative evaluation metric (see [6]), † and ‡ use different training data: Vox-Celeb [30] and VoxCeleb+ (the union of VoxCeleb and Vox-Celeb2 [2]) respectively. details and cannot bridge intra-class variations correctly.”
A Novel Distance Learning for Elastic Cross-Modal Audio-Visual Matching
Wang, R., Huang, H., Zhang, X., Ma, J., & Zheng, A. (2019) · IEEE International Conference on Multimedia & Expo Workshops (ICMEW) · doi:10.1109/icmew.2019.00-70 · Cited by 10
“• Quantitative and qualitative evaluations on benchmark dataset VoxCeleb [11] demonstrate the effectiveness of the proposed model, which yields a new state-of-theart for audio-visual matching, comparing to the other methods.”
MethodA Machine of Few Words: Interactive Speaker Recognition with Reinforcement Learning
Seurin, M., Strub, F., Preux, P., & Pietquin, O. (2020) · Interspeech (USB) · doi:10.21437/interspeech.2020-2892 · Cited by 8
Hierarchically Attending Time-Frequency and Channel Features for Improving Speaker Verification
Wang, C., Yi, J., Tao, J., Bai, Y., & Tian, Z. (2021) · International Symposium on Chinese Spoken Language Processing (ISCSLP) · doi:10.1109/iscslp49672.2021.9362054 · Cited by 7
“We have run our experiments on the VoxCeleb1 dataset [23].”
MethodInfluentialCross-lingual Speaker Verification: Evaluation on X-Vector Method
Mandalapu, H., Elbo, T. M., Ramachandra, R., & Busch, C. (2021) · Communications in computer and information science · doi:10.1007/978-3-030-71711-7_18 · Cited by 7
Speaker Embeddings by Modeling Channel-Wise Correlations
Stafylakis, T., Rohdin, J., & Burget, L. (2021) · Interspeech (USB) · doi:10.21437/interspeech.2021-1442 · Cited by 9
Squeezing Value of Cross-Domain Labels: A Decoupled Scoring Approach for Speaker Verification
Li, L., Zhang, Y., Kang, J., Zheng, T. F., & Wang, D. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9414794 · Cited by 7
Málek, J., Janský, J., Kounovský, T., Koldovský, Z., & Žďánský, J. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9414331 · Cited by 7
“The training data for the TDNN originate from the development part of the Voxceleb database [18] and the training part of the LibriSpeech corpus [19].”
MethodTime-Domain Loss Modulation Based on Overlap Ratio for Monaural Conversational Speaker Separation
Taherian, H., & Wang, D. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9413900 · Cited by 7
“VoxCeleb corpus [21, 22] and the training part of LibriSpeech dataset are used for training d-vectors.”
MethodImproved Meta-Learning Training for Speaker Verification
Chen, Y., Guo, W., & Gu, B. (2021) · Interspeech (USB) · doi:10.21437/interspeech.2021-405 · Cited by 7
Self-Supervised Metric Learning With Graph Clustering For Speaker Diarization
Singh, P., & Ganapathy, S. (2021) · IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU) · doi:10.1109/asru51503.2021.9688271 · Cited by 9
Scalable Identity-Oriented Speech Retrieval
Chen, C., Jiang, D., Peng, J., Lian, R., Li, Y., Zhang, C., Chen, L., & Fan, L. (2021) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3127520 · Cited by 8
“To mimic the real-world scenarios with multiple speech sources, we collect different public speech datasets for our evaluation, including VoxCeleb1 [31], VoxCeleb2 [32], Aishell-1 [33], Aishell-2 [34] and MAGIDATA 1 .”
MethodMulti-Channel Speaker Verification with Conv-Tasnet Based Beamformer
Mošner, L., Plchot, O., Burget, L., & Černocký, J. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9747771 · Cited by 10
Tackling Spoofing-Aware Speaker Verification with Multi-Model Fusion
Wu, H., Kang, J., Meng, L., Zhang, Y., Wu, X., Wu, Z., Lee, H.-Y., & Meng, H. M. L. (2022) · The Speaker and Language Recognition Workshop (Odyssey) · doi:10.21437/odyssey.2022-13 · Cited by 7
“Such systems can rely on state-of-the-art speaker verification models trained on large-scale datasets [4,5] and state-of-the-art spoofing countermeasure models.”
BackgroundMethodDisentangled Speaker Representation Learning via Mutual Information Minimization
Mun, S. H., Han, M. H., Kim, M., Lee, D., & Kim, N. S. (2022) · Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) · doi:10.23919/apsipaasc55919.2022.9979966 · Cited by 12
“To pre-train the front-end encoder network E θ , we employ the development set of VoxCeleb1 and VoxCeleb2 datasets [35]–[37], which consist of 1,092,009 and 148,642 utterances from 5,994 and 1,211 speakers, respectively.”
MethodKwon, Y., Heo, H. S., Huh, J., Lee, B., & Chung, J. S. (2021) · IEEE Spoken Language Technology Workshop (SLT) · doi:10.1109/slt48900.2021.9383502 · Cited by 7
“8% equal error rate on the original VoxCeleb1 test set [38].”
BackgroundSelf-citationPairwise Discriminative Neural PLDA for Speaker VerificationPreprint
Ramoji, S., Krishnan, P., Singh, P., & Ganapathy, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2001.07034 · Cited by 7
“The x-vector extractor is trained entirely using speech data extracted from combined VoxCeleb 1 [22] and VoxCeleb 2 corpora [23].”
MethodWilkinson, N., Biswas, A., Yılmaz, E., de Wet, F., van der Westhuizen, E., & Niesler, T. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2004.06480 · Cited by 7
“In our first experiment, the X-vector diarization recipe provided in the Kaldi toolkit was applied using an Xvector DNN model pre-trained on wide-band VoxCeleb data (Povey et al., 2011; Nagrani et al., 2017; Chung et al., 2018).”
MethodFew Shot Text-Independent speaker verification using 3D-CNNPreprint
Mishra, P. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.11088 · Cited by 7
Scalable Data Annotation Pipeline for High-Quality Large Speech Datasets DevelopmentPreprint
Liu, M., Zhang, C., Xing, H., Feng, C., Chen, M., Bishop, J., & Ngapo, G. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2109.01164 · Cited by 10
Understanding the Tradeoffs in Client-side Privacy for Downstream Speech TasksPreprint
Wu, P., Liang, P. P., Shi, J., Salakhutdinov, R., Watanabe, S., & Morency, L. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2101.08919 · Cited by 16
“Classifier: We use the VGGVox model, a modified version of the VGG-16 CNN, as our privacy attribute classi-fier [65, 66], slightly modifying the network by adding a ReLU activation followed by a fully-connected layer with size-2 output.”
MethodAttention-based Multimodal learning framework for Generalized Audio-Visual Deepfake Detection
Masood, M., Javed, A., & Irtaza, A. (2023) · Preprint (Research Square) · doi:10.21203/rs.3.rs-3415144/v1 · Cited by 6
HANSEN: Human and AI Spoken Text Benchmark for Authorship Analysis
Tripto, N., Uchendu, A., Le, T., Setzu, M., Giannotti, F., & Lee, D. (2023) · Findings of the Association for Computational Linguistics: EMNLP 2023 · doi:10.18653/v1/2023.findings-emnlp.916 · Cited by 6
SpeechNet: A Universal Modularized Model for Speech Processing Tasks
Chen, Y.-C., Chi, P.-H., Yang, S.-W., Chang, K.-W., Lin, J.-H., Huang, S.-F., Liu, D.-R., Liu, C.-L., Lee, C.-K., & Lee, H.-Y. (2021) · arXiv · doi:10.48550/arxiv.2105.03070 · Cited by 6
“For SC, VoxCeleb1 (Nagrani et al., 2017) contains speech uttered by 1,251 celebrities extracted from videos uploaded to YouTube.”
MethodProgressive channel fusion for more efficient TDNN on speaker verification
Zhao, Z., Li, Z., Wang, W., & Ji, X. (2024) · Speech Communication · doi:10.1016/j.specom.2024.103105 · Cited by 8
Disrupting Deepfakes: A Survey on Adversarial Perturbation Techniques and Prevention Strategies
Bagaria, U., Kumar, V., Rajesh, T., Deepak, V., & Shylaja, S. S. (2024) · International Conference on Computing and Artificial Intelligence (ICCAI) · doi:10.1145/3669754.3669799 · Cited by 10
“The survey provides a comprehensive analysis of various deep-fake generation methods, utilizing datasets such as CelebA [1], VoxCeleb [2], CelebA-HQ [3], and the DFDC dataset [4, 5].”
Metadata
- Authors
- Nagrani, A., Chung, J. S., & Zisserman, A.
- Venue
- Interspeech 2017
- Publisher
- ISCA
- Published
- 2017-08-20
- Pages
- 2616-2620
- ISSN
- 2308-457X (ISSN-L)
- OpenAlex
- W2726515241
- Cited by
- 2,210
- References
- 4
- Citations by year
- 2026: 144 · 2025: 280 · 2024: 313 · 2023: 361 · 2022: 330 · 2021: 319 · 2020: 265 · 2019: 151 · 2018: 47
- FWCI
- 56.51
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- Interspeech (USB) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Engineering and Physical Sciences Research Council (EP/M013774/1)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Signal Processing › Speech and Audio Processing
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
- Keywords
- speaker identification, speaker verification, VoxCeleb, text-independent speaker recognition, automated dataset generation, two-stream CNN, face recognition, YouTube videos
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.