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 601–700 of 1,008
Stealthy Backdoor Attack Against Speaker Recognition Using Phase-Injection Hidden Trigger
Ye, Z., Yan, D., Dong, L., Deng, J., & Yu, S. (2023) · IEEE Signal Processing Letters · doi:10.1109/lsp.2023.3293429 · Cited by 23
“The experiments are conducted on VoxCeleb1 [22] dataset.”
MethodParametric Implicit Face Representation for Audio-Driven Facial Reenactment
Huang, R., Lai, P., Qin, Y., & Li, G. (2023) · 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/cvpr52729.2023.01227 · Cited by 19
“, LRW [7], Voxceleb1 [20] and Voxceleb2 [6]) are not suitable as they either have low image quality or consist of many short (a few seconds) video clips of different speakers (e.”
BackgroundSpiking-Leaf: A Learnable Auditory Front-End for Spiking Neural Networks
Song, Z., Wu, J., Zhang, M., Shou, M. Z., & Li, H. (2024) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp48485.2024.10446789 · Cited by 19
“For speaker identification (SI), we use the Voxceleb1 dataset [29] with 153,516 utterances from 1,251 speakers, resulting in a classification task with 1,251 classes.”
MethodAV-SUPERB: A Multi-Task Evaluation Benchmark for Audio-Visual Representation Models
Tseng, Y., Berry, L., Chen, Y.-T., Chiu, I.-H., Lin, H.-H., Liu, M., Peng, P., Shih, Y.-J., Wang, H., Wu, H., et al. (2024) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp48485.2024.10445941 · Cited by 32
“…the close relation between audition and vision facilitates multimodal human perception, many audio-visual datasets have been gathered for action recognition [19, 15, 20, 21], speech recognition [22, 23, 24], speaker recognition [25, 26], and a variety of other tasks to study audio-visual learning.”
BackgroundDS-TDNN: Dual-Stream Time-Delay Neural Network With Global-Aware Filter for Speaker Verification
Li, Y., Gan, J., Lin, X., Qiu, Y., Zhan, H., & Tian, H. (2024) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2024.3402072 · Cited by 21
“Data Preparation VoxCeleb1 & 2 [40], [41], and SITW [42] are used in our experiments.”
MethodELP-Adapters: Parameter Efficient Adapter Tuning for Various Speech Processing Tasks
Inoue, N., Otake, S., Hirose, T., Ohi, M., & Kawakami, R. (2024) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2024.3434445 · Cited by 17
“The VoxCeleb1 dataset [55] is used for training and testing.”
MethodSurvey of Large Multimodal Model Datasets, Application Categories and Taxonomy
Pattnayak, P., Patel, H. L., Kumar, B., Agarwal, A., Banerjee, I., Panda, S., & Kumar, T. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-14041-8_4 · Cited by 33
Cross Attentive Pooling for Speaker Verification
Kye, S. M., Kwon, Y., & Chung, J. S. (2021) · IEEE Spoken Language Technology Workshop (SLT) · doi:10.1109/slt48900.2021.9383565 · Cited by 15
“While mainstream literature in the field have learnt speaker embeddings via the classification loss [1, 2, 3, 4, 5], such objective functions are not designed to optimise embedding similarity.”
BackgroundMethodSelf-citationListen, Look, and Find the One
Wang, X., Liu, W., Chen, J., Wang, X., Yan, C. C., & Mei, T. (2020) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3380549 · Cited by 15
Chowdhury, S. A., Durrani, N., & Ali, A. M. A. (2023) · Computer Speech & Language · doi:10.1016/j.csl.2023.101539 · Cited by 26
RSKNet-MTSP: Effective and portable deep architecture for speaker verification
Wu, Y., Guo, C., Zhao, J., Jin, X. H., & Xu, J. (2022) · Neurocomputing · doi:10.1016/j.neucom.2022.09.014 · Cited by 17
Realistic talking face animation with speech-induced head motion
Biswas, S., Sinha, S., Das, D., & Bhowmick, B. (2021) · Indian Conference on Computer Vision, Graphics and Image Processing · doi:10.1145/3490035.3490305 · Cited by 19
“We use Voxceleb1 [28], VoxCeleb2 [11], and LRS-TED [1] datasets for our experiments.”
MethodResultInfluentialMixture Representation Learning for Deep Speaker Embedding
Lin, W., & Mak, M. (2022) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2022.3153270 · Cited by 15
“The training data include the VoxCeleb1 development set and the VoxCeleb2 development set [27], [28].”
MethodRobust Speaker Verification Using Population-Based Data Augmentation
Lin, W., & Mak, M. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9746956 · Cited by 21
“The training data include the VoxCeleb1 development set and the VoxCeleb2 development set [12,13].”
MethodPhoneyTalker: An Out-of-the-Box Toolkit for Adversarial Example Attack on Speaker Recognition
Chen, M., Lü, L., Ba, Z., & Ren, K. (2022) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom48880.2022.9796934 · Cited by 15
“To train these models, we employ another large-scale corpus VoxCeleb1 [31], which contains 1,251 speakers and 153,516 utterances.”
BackgroundMethodCretois, B., Rosten, C. M., & Sethi, S. S. (2022) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.14005 · Cited by 23
“Finally, we normalized the Mel spectrograms along each frequency bin, as this step has been shown to significantly im - prove classifier performance for audio classification tasks (Nagrani et al., 2017 ).”
MethodDetecting Vocal Fatigue with Neural Embeddings
Bayerl, S. P., Wagner, D., Baumann, I., Bocklet, T., & Riedhammer, K. (2023) · Journal of Voice · doi:10.1016/j.jvoice.2023.01.012 · Cited by 21
Investigation on Neural Bandwidth Extension of Telephone Speech for Improved Speaker Recognition
Nidadavolu, P. S., Iglesias, V., Villalba, J., & Dehak, N. (2019) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2019.8682992 · Cited by 15
“WB data used in this work comprises of microphone recordings from Mixer6, SRE08 microphone speech and VoxCeleb [9] data sets.”
MethodA human data-driven interaction estimation using IoT sensors for workplace design
Hoon, J., & Cha, S. H. (2020) · Automation in Construction · doi:10.1016/j.autcon.2020.103352 · Cited by 16
Spectrogram Features for Audio and Speech Analysis
McLoughlin, I. V., Pham, L. D., Song, Y., Miao, X., Phan, H., Cai, P., Gu, Q., Nan, J., Song, H., & Soh, D. C. L. (2026) · Applied Sciences · doi:10.3390/app16020572 · Cited by 26
“Performance is generally measured by Equal Error Rate (EER) and Detection Cost Function (DCF) [142].”
Wang, J., Li, C., Zheng, A., Tang, J., & Luo, B. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3222936 · Cited by 14
“…0.1 in 600 and 1 k iterations with a maximum iteration T max of 1.2 k. Dataset: The performance of the proposed algorithm is evaluated on Voxceleb [35] and VGGFace [36] public datasets, which contain 149354 speech segments for 1225 speaker identities and 137060 face images for 1225 face…”
MethodInfluentialMotion Transformer for Unsupervised Image Animation
Tao, J., Wang, B., Ge, T., Jiang, Y., Li, W., & Duan, L. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19787-1_40 · Cited by 25
Bayesian Self-Attentive Speaker Embeddings for Text-Independent Speaker Verification
Zhu, Y., & Mak, B. K.-W. (2023) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2023.3244502 · Cited by 14
“Table IV compares the performance of our SV systems with some other systems on VoxCeleb1.”
BackgroundMethodInfluentialDepth-First Neural Architecture With Attentive Feature Fusion for Efficient Speaker Verification
Liu, B., Chen, Z., & Qian, Y. (2023) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2023.3273417 · Cited by 16
“Voxceleb1&2 [56], [57] are large-scale benchmark datasets for speaker identification and verification which contain over 6000 celebrities’ interview audio data collected from YouTube”
BackgroundLaughTalk: Expressive 3D Talking Head Generation with Laughter
Kim, S.-B., Hyun, L., Hong, D. H., Nam, S., Ju, J., & Oh, T.-H. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00628 · Cited by 28
“On the other hand, many 2D face video datasets [2,6,7,19,27,45,47] are available.”
BackgroundEnhancing Speaker Recognition Models with Noise-Resilient Feature Optimization Strategies
Chauhan, N., Isshiki, T., & Li, D. (2024) · Acoustics · doi:10.3390/acoustics6020024 · Cited by 14
“In [48], an entirely automated pipeline that leveraged computer vision techniques was employed to create voxceleb1 data from open-source media.”
MethodAshour, M. I., Abbas, A. S., Ahmed, S. R., Haydar, N., Ghaffoori, A. J., Hussain, A.-S. T., Kurdi, N. A., Al-Sarem, M., & Tawfeq, J. F. (2024) · Cognitive Models and Artificial Intelligence Conference (AICCONF) · doi:10.1145/3660853.3660938 · Cited by 14
“In addition, big-data sets have been recommended as a viable option to boost research in this subject, and such systems as VoxCeleb have been created [11], collecting valuable data for training and testing speaker recognition algorithms.”
MethodAnidjar, O. H., Marbel, R., & Yozevitch, R. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.124671 · Cited by 15
ReptoNet: A 3D Log Mel Spectrogram-Based Few-Shot Speaker Identification with Reptile Algorithm
Saritha, B., Laskar, M. A., Kirupakaran, A. M., Laskar, R. H., & Choudhury, M. (2024) · Arabian Journal for Science and Engineering · doi:10.1007/s13369-024-09426-3 · Cited by 15
“On VoxCeleb1 [41], 1251 speakers contributed 153,516 recordings for 352 h of speaking.”
Jakubec, M., Lieskovská, E., Jarina, R., Spišiak, M., & Kasák, P. (2024) · Applied Sciences · doi:10.3390/app14219981 · Cited by 24
A Benchmark for Multi-Speaker Anonymization
Miao, X., Tao, R., Zeng, C., & Wang, X. (2025) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2025.3556345 · Cited by 26
“(i) Resynthesized speech achieves DERs very similar to those of the original conversation, having almost no speaker distinctiveness loss.”
DiffusionAct: Controllable Diffusion Autoencoder for One-shot Face Reenactment
Bounareli, S., Tzelepis, C., Argyriou, V., Patras, I., & Tzimiropoulos, G. (2025) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg61629.2025.11099159 · Cited by 24
“Finally, on identity preservation (CSIM) our method remains competitive, however we argue that a higher TABLE I: Quantitative results on self reenactment task using the test set of VoxCeleb1 dataset [37].”
InfluentialLightweight speaker verification for online identification of new speakers with short segments
Vélez, I., Rascón, C., & Fuentes-Pineda, G. (2020) · Applied Soft Computing · doi:10.1016/j.asoc.2020.106704 · Cited by 14
Whosecough: In-the-Wild Cougher Verification Using Multitask Learning
Whitehill, M., Garrison, J., & Patel, S. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9053268 · Cited by 14
“To further reduce the impact of background noise, we train the speaker verification task on the Voxceleb dataset [13].”
MethodUnsupervised Training of Siamese Networks for Speaker Verification
Khan, U. M., & Hernando, J. (2020) · Interspeech (USB) · doi:10.21437/interspeech.2020-1882 · Cited by 14
“The evaluation was performed on the test partition of VoxCeleb-1 [27], which contains 40 speakers having 4,874 utterances in total.”
MethodInfluentialRobust Deep Speaker Recognition: Learning Latent Representation with Joint Angular Margin Loss
Chowdhury, L., Zunair, H., & Mohammed, N. (2020) · Applied Sciences · doi:10.3390/app10217522 · Cited by 14
“We also plan to evaluate larger datasets such as VoxCeleb [26] and VoxCeleb2 [24], which contain millions of samples from more than 6000 speakers.”
BackgroundMethodEnhancing the Perceived Emotional Intelligence of Conversational Agents through Acoustic Cues
Hu, J., Huang, Y., Hu, X., & Xu, Y. (2021) · CHI Conference Extended Abstracts on Human Factors in Computing Systems (CHI EA) · doi:10.1145/3411763.3451660 · Cited by 16
“We used a neural network model [1] trained with Voxceleb [25] for SER to get emotional input based on acoustic features in speech, Figure 1: An example dialogue of HUE.”
MethodAbraham, J. V. T., Khan, A. N., & Shahina, A. (2021) · International Journal of Speech Technology · doi:10.1007/s10772-021-09888-y · Cited by 14
“The authors in Nagrani et al. (2017) proposed CNN architecture with three variants: (i) CNN with variable length test data (ii) CNN with 3 s test data variance normalized and (iii) with no normalization.”
BackgroundMethodUnsupervised Speech Enhancement with Speech Recognition Embedding and Disentanglement Losses
Trinh, V. A., & Braun, S. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9746973 · Cited by 14
Mapping Discrete Emotions in the Dimensional Space: An Acoustic Approach
Trnka, M., Darjaa, S., Ritomský, M., Sabo, R., Rusko, M., Schaper, M., & Stelkens-Kobsch, T. H. (2021) · Electronics · doi:10.3390/electronics10232950 · Cited by 15
“Therefore, the X-vector extractor was trained on the speaker-verification databases: Vox-Celeb [34], having 1250 speakers and 150,000 utterances, and VoxCeleb2 [35] having 6000 speakers and 1.1 million utterances.”
MethodInfluentialKethireddy, R., Kadiri, S. R., & Gangashetty, S. V. (2022) · The Journal of the Acoustical Society of America · doi:10.1121/10.0009405 · Cited by 16
Discriminative speaker embedding with serialized multi-layer multi-head attention
Zhu, H., Lee, K. A., & Li, H. (2022) · Speech Communication · doi:10.1016/j.specom.2022.09.003 · Cited by 14
Cross-Age Speaker Verification: Learning Age-Invariant Speaker Embeddings
Qin, X., Li, N., Chao, W., Su, D., & Li, M. (2022) · Interspeech (USB) · doi:10.21437/interspeech.2022-648 · Cited by 26
Personality trait estimation in group discussions using multimodal analysis and speaker embedding
Mawalim, C. O., Okada, S., Nakano, Y., & Unoki, M. (2023) · Journal on Multimodal User Interfaces · doi:10.1007/s12193-023-00401-0 · Cited by 15
Conditional Injective Flows for Bayesian Imaging
Khorashadizadeh, A., Kothari, K., Salsi, L., Aghababaei-Harandi, A., de Hoop, M. V., & Dokmanić, I. (2023) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2023.3248949 · Cited by 18
Decomposition and Reorganization of Phonetic Information for Speaker Embedding Learning
Hong, Q.-B., Wu, C., & Wang, H. (2023) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2023.3267833 · Cited by 16
“The experiments were conducted on the Librispeech [44], VoxCeleb [45], [46], SITW [47], VoxMovies [58] and HI-MIA [69] datasets.”
MethodSpeech Separation with Large-Scale Self-Supervised Learning
Chen, Z., Kanda, N., Wu, J., Wu, Y., Wang, X., Yoshioka, T., Li, J., Sivasankaran, S., & Eskimez, Ş. E. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10096876 · Cited by 18
“We first shortly warmed up the SS network by training the model with artificially mixed data from VoxCeleb dataset [34, 35].”
MethodOn Learning Vocal Tract System Related Speaker Discriminative Information from Raw Signal Using CNNs
Muckenhirn, H., Magimai.-Doss, M., & Marcel, S. (2018) · Interspeech (USB) · doi:10.21437/interspeech.2018-1696 · Cited by 15
“Our future work will focus along two directions: (a) further investigations on challenging databases such as NIST SRE, VoxCeleb [9] and (b) development of i-vector based systems using the features learned by the CNNs.”
BackgroundMethodVAE-based Domain Adaptation for Speaker Verification
Wang, X., Li, L., & Wang, D. (2019) · Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) · doi:10.1109/apsipaasc47483.2019.9023015 · Cited by 14
“VoxCeleb: A large-scale free speaker database collected by University of Oxford, UK [20].”
BackgroundDisentangled Speaker and Nuisance Attribute Embedding for Robust Speaker Verification
Kang, W. H., Mun, S. H., Han, M. H., & Kim, N. S. (2020) · IEEE Access · doi:10.1109/access.2020.3012893 · Cited by 20
“• VoxCeleb1 dataset: a text-independent speaker verification corpus consisting of speech samples with 8 different emotional states [28].”
BackgroundMethodA Modulation-Domain Loss for Neural-Network-Based Real-Time Speech Enhancement
Vuong, T., Xia, Y., & Stern, R. M. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9414965 · Cited by 21
“We evaluated our system’s capabilities to improve speaker verification performance in noisy conditions by using a modified version of the VoxCeleb1 test set [27].”
MethodVector learning representation for generalized speech emotion recognition
Singkul, S., & Woraratpanya, K. (2022) · Heliyon · doi:10.1016/j.heliyon.2022.e09196 · Cited by 16
Voice Cloning: a Multi-Speaker Text-to-Speech Synthesis Approach based on Transfer LearningPreprint
Ruggiero, G., Zovato, E., Di, L., & Pollet, V. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2102.05630 · Cited by 32
“Each of them was trained using a combination of three public sets: LibriTTS [16] train-other and dev-other; VoxCeleb [17] dev and VoxCeleb2 [18] dev.”
MethodShouted and whispered speech compensation for speaker verification systems
Prieto, S., Ortega, A., López‐Espejo, I., & Lleida, E. (2022) · Digital Signal Processing · doi:10.1016/j.dsp.2022.103536 · Cited by 13
Self-Lifting: A Novel Framework for Unsupervised Voice-Face Association Learning
Chen, G., Zhang, D., Liu, T., & Du, X. (2022) · International Conference on Multimedia Retrieval · doi:10.1145/3512527.3531364 · Cited by 22
Robust Speaker Verification Using Deep Weight Space Ensemble
Lin, W., & Mak, M. (2023) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2022.3233231 · Cited by 14
“of the VoxCeleb [35] corpus (which is referred to as the base model) and used it to produce a set of speaker embeddings from English and Mandarin speech.”
MethodSelf-Supervised Training of Speaker Encoder With Multi-Modal Diverse Positive Pairs
Tao, R., Lee, K. A., Das, R. K., Hautamäki, V., & Li, H. (2023) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2023.3268568 · Cited by 18
“We evaluate the system on the VoxCeleb1 dataset [4] B. Data augmentation 1) Speech augmentation: In contrastive learning [17], data augmentation improves the diversity of training samples, thus the robustness of speaker embedding.”
DiffV2S: Diffusion-based Video-to-Speech Synthesis with Vision-guided Speaker Embedding
Choi, J., Hong, J. H., & Ro, Y. M. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00718 · Cited by 40
“It is originally trained for speaker verification task on a combination of the large speech datasets [8, 35, 37], which is trained to optimize a generalized speaker verification loss.”
Interpretable Spectrum Transformation Attacks to Speaker Recognition Systems
Yao, J., Luo, H., Qi, J., & Zhang, X.-L. (2024) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2024.3364100 · Cited by 16
“The adversarial attack experiments were conducted on the VoxCeleb [59] and LibriSpeech [60] datasets: To attack an ASV system, we arbitrarily selected 1,000 trials, including 40 speakers, from the Original-Clean trial list of Vox-Celeb1,whichincludes500targettrialsand500non-targettrials.”
MethodFSIR: Few-Shot Speaker Identification using Reptile Algorithm
Saritha, B., Kirupakaran, A. M., Laskar, R. H., & Choudhury, M. (2023) · International Conference on Computers and Devices for Communication · doi:10.1109/codec60112.2023.10466164 · Cited by 13
Priebe, D., Ghani, B., & Stowell, D. (2024) · Sensors · doi:10.3390/s24072046 · Cited by 14
“More specifically, the adjustment from 5x5 filter with a 3x3 filter in the later convolution blocks, and adjustment from the hard-swish activation function to RELU.”
MethodIncreasing Importance of Joint Analysis of Audio and Video in Computer Vision: A Survey
Shahabaz, A., & Sarkar, S. (2024) · IEEE Access · doi:10.1109/access.2024.3391817 · Cited by 18
Fan, Z., Zhang, Q., An, H., Xu, B., Xu, L., Tseng, C.-W., Peng, Y., Bejarano-Carbó, A., Abillama, P., Cao, A., et al. (2024) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2024.3410306 · Cited by 18
“We use the COCO2017 [34] dataset for FD and FR, the TIMIT speech dataset [35] for AC, and the VoxCeleb [36] dataset for CMV.”
MethodCoding Speech Through Vocal Tract Kinematics
Cho, C. J., Wu, P., Prabhune, T., Agarwal, D., & Anumanchipalli, G. K. (2024) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2024.3497655 · Cited by 50
“Both x-vector and r-vector models are trained on VoxCeleb corpus [76], [77], [78].”
MethodThe DKU System for the Speaker Recognition Task of the 2019 VOiCES from a Distance Challenge
Cai, D., Qin, X., Cai, W., & Li, M. (2019) · Interspeech (USB) · doi:10.21437/interspeech.2019-1435 · Cited by 13
“The training data includes VoxCeleb 1 [37] and VoxCeleb 2 [38].”
MethodDisentangled Speech Embeddings Using Cross-Modal Self-Supervision
Nagrani, A., Chung, J. S., Albanie, S., & Zisserman, A. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9054057 · Cited by 108
“We then extract identity embeddings for the data in the VoxCeleb1 test set (VoxCeleb1-O, 40 speakers) [4].”
MethodSelf-citationȚapu, R., Mocanu, B., & Zaharia, T. B. (2019) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2019.00313 · Cited by 15
“As recommended in [6], we modified ResNet-34 in a fully convolutional way to adapt it to 2D spectrogram inputs.”
MethodLearning Mixture Representation for Deep Speaker Embedding Using Attention
Lin, W., Mak, M., & Lu, Y. (2020) · The Speaker and Language Recognition Workshop (Odyssey) · doi:10.21437/odyssey.2020-30 · Cited by 13
“The training data includes the VoxCeleb1 development set and the VoxCeleb2 development set [18,19].”
MethodSpeaker Identification in Multi-Talker Overlapping Speech Using Neural Networks
Tran, V.-T., & Tsai, W.-H. (2020) · IEEE Access · doi:10.1109/access.2020.3009987 · Cited by 15
“The second category, e.g., in [18], [19], replaces the conventional SID classification process with a discriminating deep network in the decision making.”
BackgroundText-Independent Speaker Verification with Dual Attention Network
Li, J., & Lee, T. (2020) · Interspeech (USB) · doi:10.21437/interspeech.2020-2031 · Cited by 13
“[4] J. S. Chung, A. Nagrani, and A. Zisserman, “Voxceleb2: Deep speaker recognition,” in INTERSPEECH, 2018.”
MethodInfluentialEvaluating MFCC-based speaker identification systems with data envelopment analysis
Özcan, Z., & Kayıkçıoğlu, T. (2020) · Expert Systems with Applications · doi:10.1016/j.eswa.2020.114448 · Cited by 13
Attention Fusion for Audio-Visual Person Verification Using Multi-Scale Features
Hörmann, S., Moiz, A., Knoche, M., & Rigoll, G. (2020) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg47880.2020.00074 · Cited by 13
“Most person recognition systems rely on either images [1]–[5] or audio [6]–[12] as inputs, which are commonly extracted from a video and might become unreliable under unconstrained environments.”
BackgroundMethodResultExploring End-to-End Multi-Channel ASR with Bias Information for Meeting Transcription
Wang, X., Kanda, N., Gaur, Y., Chen, Z., Meng, Z., & Yoshioka, T. (2021) · IEEE Spoken Language Technology Workshop (SLT) · doi:10.1109/slt48900.2021.9383500 · Cited by 13
“A d-vector extractor trained on VoxCeleb Corpus [33, 34] is employed in this work [35].”
MethodCryptographic Key Derivation from Biometric Inferences for Remote Authentication
Uzun, E., Yagemann, C., Chung, S. P. H., Kolesnikov, V., & Lee, W. (2021) · ACM Asia Conference on Computer and Communications Security (ASIA CCS) · doi:10.1145/3433210.3437512 · Cited by 19
“We use VoxCeleb dataset, including 100 thousand of English utterances from 1,251 identities, to train our model [57].”
BackgroundMethodKishan, K. C., Tan, Z., Chen, L., Jin, M., Han, E., Stolcke, A., & Lee, C. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9747613 · Cited by 14
“The VoxCeleb1 dataset [23] is used for evaluation.”
MethodTarget Speech Extraction: Independent Vector Extraction Guided by Supervised Speaker Identification
Málek, J., Janský, J., Koldovský, Z., Kounovský, T., Čmejla, J., & Žďánský, J. (2022) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2022.3190739 · Cited by 13
“The training data for the FSMN and PLDA originate from the development part of the Voxceleb1 database [46] and the training part of the LibriSpeech corpus [47].”
Wu, X., Zhang, X., Feng, X., López, M. B., & Liu, L. (2022) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2022.3220040 · Cited by 21
“Besides, we also evaluate the pretrained deep models as feature encoders. pyannote-S: The pyannote.audio [61], [62] is an End-to-End generic PyanNet that is trained on Voxceleb [24] and Voxceleb2 [23] datasets.”
BackgroundMethodInfluentialECAPA-TDNN for Multi-speaker Text-to-speech Synthesis
Xue, J., Deng, Y., Han, Y., Li, Y., Sun, J., & Liang, J. (2022) · International Symposium on Chinese Spoken Language Processing (ISCSLP) · doi:10.1109/iscslp57327.2022.10037956 · Cited by 23
Context-Aware Fine-Tuning of Self-Supervised Speech Models
Shon, S., Wu, F., Kim, K., Sridhar, P., Livescu, K., & Watanabe, S. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10094687 · Cited by 13
Supervised Hierarchical Clustering Using Graph Neural Networks for Speaker Diarization
Singh, P., Kaul, A., & Ganapathy, S. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10095372 · Cited by 13
Speaker Recognition in Realistic Scenario Using Multimodal Data
Shah, S. H., Saeed, M. S., Nawaz, S., & Yousaf, M. H. (2023) · doi:10.1109/icai58407.2023.10136626 · Cited by 23
“In recent years, Convolutional Neural Networks (CNNs) have opened new paths for speaker recognition task where speech signal is converted to spectrograms to be classified with these networks [2], [3].”
InfluentialThe USTC-NERCSLIP Systems for CHiME-7 Challenge
Wan, R., He, M., Du, J., Zhou, H., Niu, S., Chen, H., Yue, Y., Yang, G., Wu, S., Sun, L., et al. (2023) · International Workshop on Speech Processing in Everyday Environments (CHiME) · doi:10.21437/chime.2023-3 · Cited by 13
Li, S., Zhang, Z. M., Tang, F., Cao, Q., Pan, H. X., & Lin, Z. (2023) · Journal of Data Science and Intelligent Systems · doi:10.47852/bonviewjdsis32021771 · Cited by 13
Learning Dense Correspondence for NeRF-Based Face Reenactment
Yang, S., Wang, W., Lan, Y., Fan, X., Peng, B., Yang, L., & Dong, J. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i7.28473 · Cited by 17
LEAP Diarization System for the Second DIHARD Challenge
Singh, P., Vardhan, H., Ganapathy, S., & Kanagasundaram, A. (2019) · Interspeech (USB) · doi:10.21437/interspeech.2019-2716 · Cited by 13
Self-Enhanced Convolutional Network for Facial Video Hallucination
Fang, C., Li, G., Han, X., & Yu, Y. (2019) · IEEE Transactions on Image Processing · doi:10.1109/tip.2019.2955640 · Cited by 13
“Practically it outperforms the second best method FRVSR by 2.7% higher PSNR and 1.0% larger SSIMvh on VoxCeleb.”
MethodInfluentialAn Empirical Analysis of Information Encoded in Disentangled Neural Speaker Representations
Peri, R., Li, H., Somandepalli, K., Jati, A., & Narayanan, S. S. (2020) · The Speaker and Language Recognition Workshop (Odyssey) · doi:10.21437/odyssey.2020-28 · Cited by 13
Privacy‐preserving speaker verification system based on binary I‐vectors
Mtibaa, A., Petrovska‐Delacrétaz, D., Boudy, J., & Hamida, A. B. (2021) · IET Biometrics · doi:10.1049/bme2.12013 · Cited by 13
“…‐ dependent and text ‐ independent speaker verification systems, we report the biometric performance evaluation of the cancel-lable i ‐ vectors system using the RSR2015 [39] text ‐ dependent database and using the VoxCeleb [40] text ‐ independent data-base for the cancellable x ‐ vectors system.”
MethodGraph Attentive Feature Aggregation for Text-Independent Speaker Verification
Shim, H.-J., Heo, J., Park, J., Lee, G.-H., & Yu, H.-J. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9746257 · Cited by 21
13 years of speaker recognition research at BUT, with longitudinal analysis of NIST SRE
Matějka, P., Plchot, O., Glembek, O., Burget, L., Rohdin, J., Zeinali, H., Mošner, L., Silnova, A., Novotný, O., Díez, M., et al. (2019) · Computer Speech & Language · doi:10.1016/j.csl.2019.101035 · Cited by 14
Adaptive Large Margin Fine-Tuning For Robust Speaker Verification
Zhang, L., Chen, Z., & Qian, Y. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10094744 · Cited by 12
“(1) Official VoxCeleb1 [15] dataset is an audio-visual large-scale dataset, containing at least 3s and average 8s utterances.”
BackgroundMethodCML-TTS: A Multilingual Dataset for Speech Synthesis in Low-Resource Languages
de Oliveira, F. S., Casanova, E., Cândido, A., Soares, A. S., & Filho, A. R. G. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-40498-6_17 · Cited by 38
Text-Inductive Graphone-Based Language Adaptation for Low-Resource Speech Synthesis
Saeki, T., Maiti, S., Li, X., Watanabe, S., Takamichi, S., & Saruwatari, H. (2024) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2024.3369537 · Cited by 16
“We used a x-vector extractor trained on VoxCeleb1 and VoxCeleb2 [57] published in SpeechBrain [58].”
MethodImproving ECAPA-TDNN Performance with Coordinate Attention
Liu, S., Song, Z., & He, L. J. (2024) · Journal of Shanghai Jiaotong University (Science) · doi:10.1007/s12204-024-2726-z · Cited by 14
Speech self-supervised representations benchmarking: A case for larger probing heads
Zaiem, S., Kemiche, Y., Parcollet, T., Essid, S., & Ravanelli, M. (2024) · Computer Speech & Language · doi:10.1016/j.csl.2024.101695 · Cited by 31
Towards Lightweight Speaker Verification via Adaptive Neural Network Quantization
Liu, B., Wang, H., & Qian, Y. (2024) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2024.3437237 · Cited by 12
“In the experiments, we evaluate the proposed methods on Voxceleb1&2 [60], [61], which are large-scale and popular benchmark datasets collected for speaker recognition tasks.”
MethodOpen Implementation and Study of Best-RQ for Speech Processing
Whetten, R., Parcollet, T., Dinarelli, M., & Estève, Y. (2024) · IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW) · doi:10.1109/icasspw62465.2024.10626364 · Cited by 28
“2 For Automatic Speaker Verification (ASV) we use Vox-celeb1 [14].”
MethodSpeech Separation With Pretrained Frontend to Minimize Domain Mismatch
Wang, W., Pan, Z., Li, X., Wang, S., & Li, H. (2024) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2024.3446242 · Cited by 18
“In [33], Han et al. extend this approach by introducing an ensemble structure combined with separation consistency training meth-ods.”
MethodBeyond Talking – Generating Holistic 3D Human Dyadic Motion for Communication
Sun, M., Xu, C., Jiang, X., Liu, Y., Sun, B., & Huang, R. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02300-7 · Cited by 25
“There exists a set of large-scale datasets for talking heads, including 2D datasets [36, 34, 17] and a 3D one [10].”
Cheng, M., & Li, M. (2025) · IEEE Transactions on Audio Speech and Language Processing · doi:10.1109/taslpro.2025.3597450 · Cited by 23
“2) MIMO-TSVAD: The audio front-end ResNet-34 is initialized by the pre-trained speaker embedding model.”
Influential
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.