sindhuhegde commited on
Commit
18335a0
·
verified ·
1 Parent(s): dfbd3b7
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -1,6 +1,6 @@
1
  # Dataset: MultiVSR
2
 
3
- We introduce a large-scale multilingual lip-reading dataset: MultiVSR. The dataset comprises a total of ~12,000 hours of video footage, covering English + 12 non-English languages. MultiVSR is a massive dataset with a huge diversity in terms of the speakers as well as languages, with approximately 1.6M video clips across 123K YouTube videos. Please check the [https://www.robots.ox.ac.uk/~vgg/research/multivsr/](website) for samples.
4
 
5
  <p align="center">
6
  <img src="dataset_teaser.gif" alt="MultiVSR Dataset Teaser">
 
1
  # Dataset: MultiVSR
2
 
3
+ We introduce a large-scale multilingual lip-reading dataset: MultiVSR. The dataset comprises a total of 12,000 hours of video footage, covering English + 12 non-English languages. MultiVSR is a massive dataset with a huge diversity in terms of the speakers as well as languages, with approximately 1.6M video clips across 123K YouTube videos. Please check the [https://www.robots.ox.ac.uk/~vgg/research/multivsr/](website) for samples.
4
 
5
  <p align="center">
6
  <img src="dataset_teaser.gif" alt="MultiVSR Dataset Teaser">