Voxceleb: large-scale speaker verification in the wild
The objective of this work is speaker recognition under noisy and unconstrained conditions. We make two key contributions. First, we introduce a very large-scale audio-visual dataset collected from open source media using a fully automated pipeline. Most existing datasets for speaker identification...
| Main Authors: | , , , |
|---|---|
| Format: | Journal article |
| Published: |
Elsevier
2019
|
| _version_ | 1826307138404745216 |
|---|---|
| author | Nagrani, A Chung, JS Xie, W Zisserman, A |
| author_facet | Nagrani, A Chung, JS Xie, W Zisserman, A |
| author_sort | Nagrani, A |
| collection | OXFORD |
| description | The objective of this work is speaker recognition under noisy and unconstrained conditions. We make two key contributions. First, we introduce a very large-scale audio-visual dataset collected from open source media using a fully automated pipeline. Most existing datasets for speaker identification contain samples obtained under quite constrained conditions, and usually require manual annotations, hence are limited in size. We propose a 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 contains over a million ‘real-world’ utterances from over 6000 speakers. This is several times larger than any publicly available speaker recognition dataset. Second, we develop and compare different CNN architectures with various aggregation methods and training loss functions that can effectively recognise identities from voice under various conditions. The models trained on our dataset surpass the performance of previous works by a significant margin. |
| first_indexed | 2024-03-07T06:58:27Z |
| format | Journal article |
| id | oxford-uuid:feee8433-55c1-43d9-a808-f3832b3be0d1 |
| institution | University of Oxford |
| last_indexed | 2024-03-07T06:58:27Z |
| publishDate | 2019 |
| publisher | Elsevier |
| record_format | dspace |
| spelling | oxford-uuid:feee8433-55c1-43d9-a808-f3832b3be0d12022-03-27T13:40:31ZVoxceleb: large-scale speaker verification in the wildJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:feee8433-55c1-43d9-a808-f3832b3be0d1Symplectic Elements at OxfordElsevier2019Nagrani, AChung, JSXie, WZisserman, AThe objective of this work is speaker recognition under noisy and unconstrained conditions. We make two key contributions. First, we introduce a very large-scale audio-visual dataset collected from open source media using a fully automated pipeline. Most existing datasets for speaker identification contain samples obtained under quite constrained conditions, and usually require manual annotations, hence are limited in size. We propose a 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 contains over a million ‘real-world’ utterances from over 6000 speakers. This is several times larger than any publicly available speaker recognition dataset. Second, we develop and compare different CNN architectures with various aggregation methods and training loss functions that can effectively recognise identities from voice under various conditions. The models trained on our dataset surpass the performance of previous works by a significant margin. |
| spellingShingle | Nagrani, A Chung, JS Xie, W Zisserman, A Voxceleb: large-scale speaker verification in the wild |
| title | Voxceleb: large-scale speaker verification in the wild |
| title_full | Voxceleb: large-scale speaker verification in the wild |
| title_fullStr | Voxceleb: large-scale speaker verification in the wild |
| title_full_unstemmed | Voxceleb: large-scale speaker verification in the wild |
| title_short | Voxceleb: large-scale speaker verification in the wild |
| title_sort | voxceleb large scale speaker verification in the wild |
| work_keys_str_mv | AT nagrania voxceleblargescalespeakerverificationinthewild AT chungjs voxceleblargescalespeakerverificationinthewild AT xiew voxceleblargescalespeakerverificationinthewild AT zissermana voxceleblargescalespeakerverificationinthewild |