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 501–600 of 1,008
Speaker recognition using PCA-based feature transformation
Ahmed, A. I., Chiverton, J. P., Ndzi, D. L., & Becerra, V. M. (2019) · Speech Communication · doi:10.1016/j.specom.2019.04.001 · Cited by 20
Saleem, S., Subhan, F., Naseer, N., Bais, A., & Imtiaz, A. (2020) · Forensic Science International Digital Investigation · doi:10.1016/j.fsidi.2020.300982 · Cited by 24
Exploring Algorithmic Fairness in Deep Speaker Verification
Fenu, G., Lafhouli, H., & Marras, M. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58811-3_6 · Cited by 25
“The most prominent examples include d-Vectors [34], c-Vectors [8], x-Vectors [31], VGGVox-Vectors [26] and ResNet-Vectors [9].”
TACR-Net: Editing on Deep Video and Voice Portraits
Song, L., Liu, B., Yin, G., Dong, X., Zhang, Y., & Bai, J.-X. (2021) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3474085.3475196 · Cited by 24
“Inspired by the Audio-ID Removing Network (AIR) in EBT [29], we apply pre-trained speaker identity network VGGVox [5, 22] on the source voice and converted”
MethodNon-Contrastive Self-Supervised Learning for Utterance-Level Information Extraction From Speech
Cho, J., Villalba, J., Moro-Velázquez, L., & Dehak, N. (2022) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2022.3197315 · Cited by 21
“VoxCeleb subset: We used VoxCeleb1 [41], VoxcSRC-21 1 val and test trials.”
MethodWang, Z., Wu, S., Chen, H., He, M., Du, J., Lee, C., Chen, J., Watanabe, S., Siniscalchi, S. M., Scharenborg, O., et al. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10094836 · Cited by 37
“In Track 1, external audio data can be used to train the AVSD model, such as VoxCeleb 1, 2 [20, 21], CN-Celeb [22], and other public datasets.”
MethodA Deep Network for Arousal-Valence Emotion Prediction with Acoustic-Visual CuesPreprint
Peng, S., Zhang, L., Ban, Y., Fang, M., & WINKLER, S. S. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1805.00638 · Cited by 20
“Similar to [6], spectrograms are then calculated every 10ms with a sliding hamming window of width 25ms and 512-point FFT.”
MethodRobust and Discriminative Speaker Embedding via Intra-Class Distance Variance Regularization
Le, N., & Odobez, J. (2018) · Interspeech (USB) · doi:10.21437/interspeech.2018-1685 · Cited by 21
“To overcome these challenges, one can employ effective sampling strategies [10, 11] or training embedding networks on top of pretrained classification models [5, 9].”
BackgroundMethodInfluentialSpeaker Attentive Speech Emotion Recognition
Veillon, C. L. M., Obin, N., & Röebel, A. (2021) · Interspeech (USB) · doi:10.21437/interspeech.2021-573 · Cited by 26
Y-Vector: Multiscale Waveform Encoder for Speaker Embedding
Zhu, G., Jiang, F., & Duan, Z. (2021) · Interspeech (USB) · doi:10.21437/interspeech.2021-1707 · Cited by 26
“In our experiment, we use the VoxCeleb corpora [24, 25] to train and evaluate our system and comparison systems for speaker verification.”
MethodMulti-Scale Speaker Embedding-Based Graph Attention Networks For Speaker Diarisation
Kwon, Y., Heo, H.-S., Jung, J.-W., Kim, Y. J., Lee, B., & Chung, J. S. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9747450 · Cited by 19
Self-citationTowards Lightweight Applications: Asymmetric Enroll-Verify Structure for Speaker Verification
Li, Q., Yang, L., Wang, X., Qin, X., Wang, J., & Li, M. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9746247 · Cited by 23
Speaker anonymization by modifying fundamental frequency and x-vector singular value
Mawalim, C. O., Galajit, K., Karnjana, J., Kidani, S., & Unoki, M. (2021) · Computer Speech & Language · doi:10.1016/j.csl.2021.101326 · Cited by 29
Incorporating Visual Information in Audio Based Self-Supervised Speaker Recognition
Cai, D., Wang, W., & Li, M. (2022) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2022.3162078 · Cited by 24
Predicting Personalized Head Movement From Short Video and Speech Signal
Yi, R., Ye, Z., Sun, Z., Zhang, J., Zhang, G., Wan, P., Bao, H., & Liu, Y. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3207606 · Cited by 25
Arezzo, A., & Berretti, S. (2022) · ACM International Conference on Multimedia in Asia (MMAsia) · doi:10.1145/3551626.3564937 · Cited by 24
“This is basically a CNN based on the Visual Geometry Group (VGG)-M [5] network trained for the speaker recognition task starting from the spectrograms of the speech signal of the Vox Celeb dataset [24].”
MethodInfluentialA Comprehensive Review on Speaker Recognition
Saritha, B., Laskar, M. A., & Laskar, R. H. (2022) · Signals and communication technology · doi:10.1007/978-3-031-18444-4_1 · Cited by 20
Kalman-Inspired Feature Propagation for Video Face Super-Resolution
Feng, R., Li, C., & Loy, C. C. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73347-5_12 · Cited by 33
You, L., Guo, W., Dai, L.-R., & Du, J. (2019) · Interspeech (USB) · doi:10.21437/interspeech.2019-2264 · Cited by 23
“We use both the Voxceleb1 and Voxceleb2 datasets [19, 20] as the training set for VOiCES Challenge.”
MethodNEC-TT System for Mixed-Bandwidth and Multi-Domain Speaker Recognition
Lee, K. A., Yamamoto, H., Okabe, K., Wang, Q., Guo, L., Koshinaka, T., Zhang, J., & Shinoda, K. (2019) · Computer Speech & Language · doi:10.1016/j.csl.2019.101033 · Cited by 21
Statistics Pooling Time Delay Neural Network Based on X-Vector for Speaker Verification
Hong, Q.-B., Wu, C., Wang, H., & Huang, C.-L. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9054350 · Cited by 19
“We found that stats-vector with score fusion can significantly improve the speaker verification performance on VoxCeleb1 dataset.”
MethodInfluentialMultimodal Speaker Diarization of Real-World Meetings Using D-Vectors With Spatial Features
Kang, W., Roy, B., & Chow, W. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9053122 · Cited by 19
“Our model was trained from scratch on the development sets of the VoxCeleb [19] and VoxCeleb2 [20] datasets, which amount to approximately 7K speakers and 1.”
MethodDilated residual networks with multi-level attention for speaker verification
Wu, Y., Guo, C., Gao, H., Xu, J., & Bai, G. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.06.079 · Cited by 19
D-MONA: A dilated mixed-order non-local attention network for speaker and language recognition
Miao, X., McLoughlin, I. V., Wang, W., & Zhang, P. (2021) · Neural Networks · doi:10.1016/j.neunet.2021.03.014 · Cited by 19
TwoStreamVAN: Improving Motion Modeling in Video Generation
Sun, X., Xu, H., & Saenko, K. (2020) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv45572.2020.9093557 · Cited by 24
“To extensively test the ability of our model to generate real-world videos, we experiment on a new large-scale speech dataset VoxCeleb [32].”
MethodFederated Learning of User Authentication ModelsPreprint
Hosseini, H., Yun, S., Park, H., Louizos, C., Soriaga, J. B., & Welling, M. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.04618 · Cited by 23
“The models are trained with data of 658 users (out of 1, 251 users) of the VoxCeleb dataset (Nagrani et al., 2017).”
MethodInfluentialUnsupervised Representation Learning for Speaker Recognition via Contrastive Equilibrium LearningPreprint
Mun, S. H., Kang, W. H., Han, M. H., & Kim, N. S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.11433 · Cited by 19
“[1] VoxCeleb1 PLDA 8.”
BackgroundMethodArtificial intelligence–powered 3D analysis of video-based caregiver-child interactions
Weng, Z., Bravo-Sánchez, L., Wang, Z., Howard, C., Xenochristou, M., Meister, N., Kanazawa, A., Milstein, A. S., Bergelson, E., Humphreys, K. L., et al. (2025) · Science Advances · doi:10.1126/sciadv.adp4422 · Cited by 29
Optimizing Bayesian Hmm Based X-Vector Clustering for the Second Dihard Speech Diarization Challenge
Díez, M., Burget, L., Landini, F., Wang, S., & Černocký, H. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9053982 · Cited by 18
“It is trained for speaker classification on VoxCeleb [13] training and VoxCeleb2 [16] development data with data augmentation, amounting to 6 million utterances from 7146 speakers.”
MethodFRILL: A Non-Semantic Speech Embedding for Mobile Devices
Peplinski, J., Shor, J., Joglekar, S., Garrison, J., & Patel, S. (2021) · Interspeech (USB) · doi:10.21437/interspeech.2021-2070 · Cited by 29
A Real-Time Speaker Diarization System Based on Spatial Spectrum
Zheng, S., Huang, W., Wang, X., Suo, H., Feng, J., & Yan, Z. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9413544 · Cited by 18
“For train set we use VoxCeleb dataset [18], since we have not collected enough labelled data from our microphone-array design to train a speaker embedding network.”
MethodGANBA: Generative Adversarial Network for Biometric Anti-Spoofing
Gomez-Alanis, A., González, J. A., & Peinado, A. M. (2022) · Applied Sciences · doi:10.3390/app12031454 · Cited by 27
“To train the TDNN [21] of the ASV system as an x-vector features extractor, we also employed the Voxceleb1 database [44] which contains more than 1000 speakers.”
MethodZhang, Z., & Ding, Y. (2022) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3503161.3548330 · Cited by 23
“The final loss 𝐿 is written as where 𝜆 𝑝𝑒𝑟𝑐 represents the weight of perceptual loss and we set 𝑝𝑒𝑟𝑐 = 5.”
Lambamo, W., Srinivasagan, R., & Jifara, W. (2022) · Applied Sciences · doi:10.3390/app13010569 · Cited by 21
Experimental Case Study of Self-Supervised Learning for Voice Spoofing Detection
Lee, Y., Kim, N., Jeong, J., & Kwak, I. (2023) · IEEE Access · doi:10.1109/access.2023.3254880 · Cited by 25
“VoxCeleb [38] was used for pre-training.”
Parameter Efficient Transfer Learning for Various Speech Processing Tasks
Otake, S., Kawakami, R., & Inoue, N. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10096311 · Cited by 19
“For ASV, the VoxCeleb1 dataset [22] is used for evaluation.”
MethodPretraining Conformer with ASR for Speaker Verification
Cai, D., Wang, W., Li, M., Xia, R., & Huang, C. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10096659 · Cited by 25
“The experiments are conducted on VoxCeleb [30, 31].”
Analyzing lower half facial gestures for lip reading applications: Survey on vision techniques
Seegehalli, P. J., & Niranjana Krupa, B. (2023) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2023.103738 · Cited by 20
VIVE3D: Viewpoint-Independent Video Editing using 3D-Aware GANs
Frühstück, A., Sarafianos, N., Xu, Y., Wonka, P., & Tung, T. (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/cvpr52729.2023.00432 · Cited by 28
Zheng, Q., Chen, Z., Wang, Z., Liu, H., & Lin, M. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.123004 · Cited by 27
ESPnet-Codec: Comprehensive Training and Evaluation of Neural Codecs For Audio, Music, and Speech
Shi, J., Tian, J., Wu, Y., Jung, J.-W., Yip, J. Q., Masuyama, Y., Chen, W., Wu, Y., Tang, Y., Baali, M., et al. (2024) · IEEE Spoken Language Technology Workshop (SLT) · doi:10.1109/slt61566.2024.10832289 · Cited by 63
“Downstream Integration: SPK Setups : Our experiments for the SPK task utilize the VoxCeleb1 [100] corpus.”
DEEP-HEAR: A Multimodal Subtitle Positioning System Dedicated to Deaf and Hearing-Impaired People
Țapu, R., Mocanu, B., & Zaharia, T. B. (2019) · IEEE Access · doi:10.1109/access.2019.2925806 · Cited by 20
“As recommended in [25], we have modified the network layers architecture in order to adapt it to spectrograms input and introduced a batch normalization stage before computing ReLUs.”
A Unified Deep Learning Framework for Short-Duration Speaker Verification in Adverse Environments
Jung, Y., Choi, Y., Lim, H., & Kim, H. (2020) · IEEE Access · doi:10.1109/access.2020.3025941 · Cited by 18
“using feature maps at multiple time-frequency scales, achieving state-of-the-art performance on VoxCeleb [31].”
BackgroundMethodInfluentialCombination of deep speaker embeddings for diarisation
Sun, G., Zhang, C., & Woodland, P. C. (2021) · Neural Networks · doi:10.1016/j.neunet.2021.04.020 · Cited by 19
Liu, Y., Song, Y., McLoughlin, I. V., Liu, L., & Dai, L.-R. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9413421 · Cited by 18
Speaker Recognition with ResNet and VGG Networks
Jakubec, M., Lieskovská, E., & Jarina, R. (2021) · International Conference Radioelektronika · doi:10.1109/radioelektronika52220.2021.9420202 · Cited by 18
Triantafyllopoulos, A., & Schuller, B. W. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9414896 · Cited by 32
Cross-Modal Perceptionist: Can Face Geometry be Gleaned from Voices?
Wu, C.-Y., Hsu, C.-C., & Neumann, U. (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/cvpr52688.2022.01020 · Cited by 25
“Most large voice datasets, such as Voxceleb [30], only contain speech for celebrities, and most large face datasets, such as VGGFace [37], only consist of publicly scraped face images.”
BackgroundSelf-Supervised Representation Learning With Path Integral Clustering for Speaker Diarization
Singh, P., & Ganapathy, S. (2021) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2021.3075100 · Cited by 23
“For training the x-vector extractor and the PLDA model in AMI experiments, we trained the TDNN model using 16 kHz data from VoxCeleb-1 [51] and VoxCeleb-2 [52] datasets containing 7323 speakers.”
MethodLarge-scale multilingual audio visual dubbingPreprint
Yang, Y., Shillingford, B., Assael, Y., Wang, M., Liu, W., Chen, Y., Zhang, Y., Sezener, E., Cobo, L. C., Denil, M., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.03530 · Cited by 38
“Datasets Some of the most commonly used datasets in the literature are GRID [10], LRW [13], VoxCeleb[15] and VoxCeleb2[49].”
MethodEgo4D: Around the World in 3,600 Hours of Egocentric Video
Grauman, K. L., Westbury, A., Byrne, E., Cartillier, V., Chavis, Z. Q., Furnari, A., Girdhar, R., Hamburger, J., Jiang, H., Kukreja, D., et al. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3381075 · Cited by 31
CAP4D: Creating Animatable 4D Portrait Avatars with Morphable Multi-View Diffusion Models
Taubner, F., Zhang, R., Tuli, M., & Lindell, D. B. (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/cvpr52734.2025.00501 · Cited by 71
“To address this limitation, recent techniques leverage large datasets of 2D portrait images [45, 65, 92] and 3D scans [20, 33, 99, 116] to train diffusion models that capture robust priors on human appearance, enabling the reconstruction of 2D [85], 3D [36, 68], or 4D avatars [18] from a single…”
Learning Lip-Based Audio-Visual Speaker Embeddings with AV-HuBERT
Shi, B., Mohamed, A., & Hsu, W.-N. (2022) · Interspeech (USB) · doi:10.21437/interspeech.2022-885 · Cited by 26
Speaker Recognition through Deep Learning Techniques: A Comprehensive Review and Research Challenges
Shome, N., Sarkar, A., Ghosh, A. K., Laskar, R. H., & Kashyap, R. (2023) · Periodica Polytechnica Electrical Engineering and Computer Science · doi:10.3311/ppee.20971 · Cited by 21
Exploring Binary Classification Loss for Speaker Verification
Han, B., Chen, Z., & Qian, Y. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10094954 · Cited by 17
“The evaluation trials in our experiment include three cleaned version trials Vox1-O, Vox1-E and Vox1-H constructed from 1251 speakers in Vox-celeb1 [17].”
MethodThe effect of speech pathology on automatic speaker verification: a large-scale study
Arasteh, S. T., Weise, T., Schuster, M. E., Nöth, E., Maier, A., & Yang, S. H. (2023) · Scientific Reports · doi:10.1038/s41598-023-47711-7 · Cited by 25
ToonTalker: Cross-Domain Face Reenactment
Gong, Y., Zhang, Y., Cun, X., Yin, F., Fan, Y., Wang, X., Wu, B., & Yang, Y. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00707 · Cited by 31
“In the first stage, we trained the model on the training set of VoxCeleb for 450K iterations, where the batch size is 64 and the optimizer is Adam with an initial learning rate of 5 × 10−4.”
MethodInfluentialComparison of Modern Deep Learning Models for Speaker Verification
Brydinskyi, V., Khoma, Y., Sabodashko, D., Podpora, M., Khoma, V., Konovalov, A., & Kostiak, M. (2024) · Applied Sciences · doi:10.3390/app14041329 · Cited by 29
CVTHead: One-shot Controllable Head Avatar with Vertex-feature Transformer
Ma, H., Zhang, T., Sun, S., Yan, X., Han, K., & Xie, X. S. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00602 · Cited by 42
“Dataset For a fair comparison with previous works, we conduct experiments on VoxCeleb1 [42] and VoxCeleb2 [10].”
MethodUnsupervised Representation Learning of Speech for Dialect Identification
Shon, S., Hsu, W.-N., & Glass, J. R. (2018) · IEEE Spoken Language Technology Workshop (SLT) · doi:10.1109/slt.2018.8639650 · Cited by 17
Unsupervised Feature Enhancement for Speaker Verification
Nidadavolu, P. S., Kataria, S., Villalba, J., García-Perera, P., & Dehak, N. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9053823 · Cited by 21
“The files from the same YouTube video of VoxCeleb1 [14] and Voxceleb2 [15] were concatenated, denoted as voxcelebcat , to obtain longer audio sequences.”
MethodThe Sound of My Voice: Speaker Representation Loss for Target Voice Separation
Mun, S., Choe, S., Huh, J., & Chung, J. S. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9053521 · Cited by 17
Self-citationImproved Active Speaker Detection based on Optical Flow
Huang, C. X., & Koishida, K. (2020) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw50498.2020.00483 · Cited by 18
“, Columbia [4] and VoxCeleb [20]), the AVA-ActiveSpeaker dataset has much more labeled data, including 3.”
BackgroundInfluentialN-HANS: A neural network-based toolkit for in-the-wild audio enhancement
Liu, S., Keren, G., Parada‐Cabaleiro, E., & Schuller, B. W. (2021) · Multimedia Tools and Applications · doi:10.1007/s11042-021-11080-y · Cited by 19
Graph-Based Label Propagation for Semi-Supervised Speaker Identification
Chen, L., Ravichandran, V., & Stolcke, A. (2021) · Interspeech (USB) · doi:10.21437/interspeech.2021-1209 · Cited by 21
“For example, VGG-M [35], VGGVox [24], AutoSpeech [25], Magneto [22] all utilize CNN-based backbone networks to learn speaker embeddings from pre-processed spectrograms of utterances.”
BackgroundMethodSerialized Multi-Layer Multi-Head Attention for Neural Speaker Embedding
Zhu, H., Lee, K. A., & Li, H. (2021) · Interspeech (USB) · doi:10.21437/interspeech.2021-2210 · Cited by 19
Combining Attention with Flow for Person Image Synthesis
Ren, Y., Wu, Y., Li, T. H., Liu, S., & Li, G. (2021) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3474085.3475256 · Cited by 17
“The VoxCeleb dataset [22] which contains 22496 talking-head videos is used for training and testing.”
MethodCross-Modal Audio-Visual Co-Learning for Text-Independent Speaker Verification
Liu, M., Lee, K. A., Wang, L., Zhang, H., Zeng, C., & Dang, J. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10095883 · Cited by 21
“As shown in Table 1, we compile an audio-visual lip biometrics dataset from LRS3 [26], LombardGrid [27], Grid [28], VoxCeleb1 [22], VoxCeleb2 [3].”
Emotion Recognition in Speech Using Cross-Modal Transfer in the WildPreprint
Albanie, S., Nagrani, A., Vedaldi, A., & Zisserman, A. (2018) · Oxford University Research Archive (ORA) (University of Oxford) · doi:10.48550/arxiv.1808.05561 · Cited by 22
“VoxCeleb [50] video dataset – this is a large-scale dataset of emotionunlabelled speaking face-tracks obtained in the wild (section 4); (iii) we transfer supervision across modalities from faces to a speech, and then train a speech emotion recognition model using speaking facetracks (section 5); and, (iv) we demonstra…”
BackgroundMethodInfluentialSelf-citationTurner, H., Lovisotto, G., & Martinović, I. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.13457 · Cited by 29
“We train and reuse the models in the exact same way as the baseline for this, with the exception that we use the VoxCeleb1 [14] and VoxCeleb2 [15] datasets in our pool of speaker X-vectors, in addition to the LibriTTS [16] train-other-500 dataset.”
MethodEfficient and robust temporal processing with neural oscillations modulated spiking neural networks
Yan, Y., Qu, Y., Wu, Y., Liu, H., Zhang, M., Li, H., Tan, K. C., & Wu, J. (2025) · Nature Communications · doi:10.1038/s41467-025-63771-x · Cited by 28
“…29 , speech recognition on Spiking Heidelberg Digits (SHD) 30 and Google Speech Commands (GSC) 31 , bio-signals recognition on Electro-cardiogram (ECG) 32 , speaker identi fi cation on VoxCeleb1 33 , language modeling on Penn Tree Bank (PTB) 34 , and event stream recognition on DVS-Gesture 35 .”
InfluentialSub-Band Convolutional Neural Networks for Small-Footprint Spoken Term Classification
Kao, C.-C., Sun, M., Gao, Y., Vitaladevuni, S. N. P., & Wang, C. (2019) · Interspeech (USB) · doi:10.21437/interspeech.2019-1766 · Cited by 17
Time-Domain Speaker Verification Using Temporal Convolutional Networks
Han, S., Byun, J., & Shin, J. W. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9414765 · Cited by 16
“To evaluate the performance of the proposed model, we used the VoxCeleb1 dataset [18], of which the training set contains 148,642 utterances from 1,211 speakers for text-independent speaker verification and the test set includes 4,874 utterances spoken by 40 speakers.”
MethodA neural network architecture optimizer based on DARTS and generative adversarial learning
Zhang, T., Waqas, M., Shen, H., Liu, Z.-Y., Zhang, X., Li, Y.-J., Halim, Z., & Chen, S. (2021) · Information Sciences · doi:10.1016/j.ins.2021.09.041 · Cited by 15
“Then, we load this new model to train with the original dataset while keeping the hyper-parameters, such as learning rate, the batch size, dropout, unchanged to compare with the network before pruning.”
Qin, X., Cai, D., & Li, M. (2022) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2022.3212834 · Cited by 19
“1&2: The VoxCeleb 1&2 [70], [71] are widely used large-scale”
MethodConvolution-Based Channel-Frequency Attention for Text-Independent Speaker Verification
Li, J., Tian, Y., & Lee, T. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10095415 · Cited by 18
“Deep neural network (DNN) based speaker embeddings such as X-vector and r-vector [4–6] showed superior performance to the factor analysis-based I-vector [7].”
MethodSaeki, T., Maiti, S., Li, X., Watanabe, S., Takamichi, S., & Saruwatari, H. (2023) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2023/575 · Cited by 25
StyleTalk++: A Unified Framework for Controlling the Speaking Styles of Talking Heads
Wang, S., Ma, Y., Ding, Y., Hu, Z., Fan, C., Lv, T., Deng, Z., & Yu, X. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3357808 · Cited by 27
“Inthispaper, wetrainedanoff-the-shelfimagerendereronthreetalkingface datasets,namelyHDTF[6],VoxCeleb[60],andMEAD[8].”
MethodKEBR: Knowledge Enhanced Self-Supervised Balanced Representation for Multimodal Sentiment Analysis
Zhu, A., Hu, M. Q., Wang, X., Yang, J., Tang, Y., & Ren, F. (2024) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3664647.3681163 · Cited by 36
“To predict the sentiment value of the masked word, we employ a multilayer perceptron (MLP) with a ReLU activation function acting as a classifier.”
Shahzad, S. A., Hashmi, A., Peng, Y., Tsao, Y., & Wang, H. (2025) · IEEE Transactions on Human-Machine Systems · doi:10.1109/thms.2025.3618409 · Cited by 31
“The statistics of the two dataset are shown in Table I.”
Analysis of Deep Feature Loss Based Enhancement for Speaker Verification
Kataria, S., Nidadavolu, P. S., Villalba, J., & Dehak, N. (2020) · The Speaker and Language Recognition Workshop (Odyssey) · doi:10.21437/odyssey.2020-66 · Cited by 16
A Comparison of Metric Learning Loss Functions for End-To-End Speaker Verification
Coria, J. M., Bredin, H., Ghannay, S., & Rosset, S. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-59430-5_11 · Cited by 17
Channel Adversarial Training for Speaker Verification and Diarization
Luu, C., Bell, P., & Renals, S. J. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9053323 · Cited by 16
Paralinguistic Privacy Protection at the Edge
Aloufi, R., Haddadi, H., & Boyle, D. (2022) · ACM Transactions on Privacy and Security · doi:10.1145/3570161 · Cited by 17
“The VoxCeleb dataset [55] contains over 100,000 utterances from 1,251 celebrities, extracted from videos uploaded to YouTube.”
BackgroundWang, Z., Xia, W., & Hansen, J. H. L. (2020) · Interspeech (USB) · doi:10.21437/interspeech.2020-2738 · Cited by 19
You're Not You When You're Angry: Robust Emotion Features Emerge by Recognizing Speakers
Aldeneh, Z., & Provost, E. M. (2021) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2021.3086050 · Cited by 18
“Some common neural representations introduced in the literature include the d-vector and x-vector representations [9], [17], [17], [18], [19].”
BackgroundInfluentialMulti-Modality Deep Restoration of Extremely Compressed Face Videos
Zhang, X., & Wu, X. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3157388 · Cited by 26
Improving Fairness in Speaker Verification via Group-Adapted Fusion Network
Shen, H., Yang, Y., Sun, G., Langman, R., Han, E., Droppo, J., & Stolcke, A. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9747384 · Cited by 30
“VoxCeleb1-F strictly controls for the presence of positive and negative trials with same or different genders, as shown in Table 2.”
BackgroundMethodInfluentialAugmentation Adversarial Training for Self-Supervised Speaker Representation Learning
Kang, J., Huh, J., Heo, H. S., & Chung, J. S. (2022) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2022.3200915 · Cited by 23
“The semi-supervised models outperform the self-supervised counterparts by a large margin, and also show strong performance compared to the model trained us-ing full supervision on VoxCeleb1 alone.”
BackgroundMethodInfluentialSelf-citationDecorrelating Feature Spaces for Learning General-Purpose Audio Representations
Ghosh, S., Seth, A., & Umesh, S. (2022) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2022.3202093 · Cited by 17
“For speaker identification, we resort to two commonly used datasets, namely, LibriSpeech (LBS) [36] and VoxCeleb [34].”
MethodEmoMV: Affective music-video correspondence learning datasets for classification and retrieval
Thao, H. T. P., Roig, G., & Herremans, D. (2022) · Information Fusion · doi:10.1016/j.inffus.2022.10.002 · Cited by 26
A Comprehensive Study on Self-Supervised Distillation for Speaker Representation Learning
Chen, Z., Qian, Y., Han, B., Qian, Y., & Zeng, M. (2023) · IEEE Spoken Language Technology Workshop (SLT) · doi:10.1109/slt54892.2023.10022470 · Cited by 24
“We evaluate our system on three different evaluation trials Vox1-O, Vox1-E, and Vox1-H constructed from Voxceleb1 dataset [25].”
MethodBird, J. J., Faria, D. R., Premebida, C., Ekárt, A., & Ayrosa, P. P. S. (2020) · IEEE International Conference on Autonomous Robot Systems and Competitions · doi:10.1109/icarsc49921.2020.9096166 · Cited by 16
Joint gender and age estimation based on speech signals using x-vectors and transfer learningPreprint
Kwaśny, D., & Hemmerling, D. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.01551 · Cited by 19
“The VoxCeleb1 dataset was created primarily with the aim of accelerating research in the field of speaker identification and verification.”
MethodLu, K.-H., Chen, Z., Fu, S., Yang, C.-H. H., Huang, S.-F., Yang, C.-K., Yu, C.-M., Chen, C., Chen, W.-C., Huang, C.-Y., et al. (2026) · IEEE Transactions on Audio Speech and Language Processing · doi:10.1109/taslpro.2026.3675792 · Cited by 82
“This allows the model to reason more effectively about context, emotion, and acoustic scenes, potentially enhancing its performance on tasks that rely on paralinguistic and contextual cues.”
Adapting Speaker Embeddings for Speaker Diarisation
Kwon, Y., Jung, J.-W., Heo, H.-S., Kim, Y. J., Lee, B., & Chung, J. S. (2021) · Interspeech (USB) · doi:10.21437/interspeech.2021-448 · Cited by 21
Self-citationZhang, X., Zhang, X., Liu, W., Zou, X., Sun, M., & Zhao, J. C. (2022) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2022.105469 · Cited by 20
Cai, D., Cai, Z., & Li, M. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10096733 · Cited by 26
“We use LibriSpeech [35] as the source speech dataset and VoxCeleb [36, 37] as the target speech dataset.”
MethodInfluentialNoise-Disentanglement Metric Learning for Robust Speaker Verification
Sun, Y., Zhang, H., Wang, L., Lee, K. A., Liu, M., & Dang, J. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10096848 · Cited by 24
“Our proposed noise-disentanglement metric learning (NDML) is evaluated on the testset of VoxCeleb1 [20].”
MethodVoicePM: A Robust Privacy Measurement on Voice Anonymity
Zhang, S., Li, Z., & Das, A. (2023) · ACM Conference on Security and Privacy in Wireless and Mobile Networks (WiSec) · doi:10.1145/3558482.3590175 · Cited by 24
“This model was trained using the Voxceleb1 [43], and Voxceleb2 [42] datasets.”
Method
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.