LRS3

The LRS3 dataset is a large-scale dataset for visual speech recognition. It consists of thousands of spoken sentences from TED videos.

Data and Resources

Original MetadataJSON
The json representation of the dataset with its distributions based on DCAT.
Explore
- Preview
- Download

Triantafyllos Afouras, Joon Son Chung, Andrew Zisserman (2024). Dataset: LRS3. https://doi.org/10.57702/i0k53x2i

DOI retrieved: December 16, 2024

Field	Value
Created	December 16, 2024
Last update	December 16, 2024
Defined In	https://doi.org/10.48550/arXiv.2012.10852
Citation	https://doi.org/10.48550/arXiv.2303.08670 https://doi.org/10.48550/arXiv.2302.13700 https://doi.org/10.48550/arXiv.2404.02098 https://doi.org/10.1109/ICASSP49357.2023.10096889 https://doi.org/10.48550/arXiv.2306.17005 https://doi.org/10.48550/arXiv.1907.04975 https://doi.org/10.48550/arXiv.2210.07055 https://doi.org/10.48550/arXiv.2305.08293
Author	Triantafyllos Afouras
More Authors	Joon Son Chung Andrew Zisserman
Homepage	https://arxiv.org/abs/1809.00496