Datasets:

Modalities:
Audio
Text
Formats:
webdataset
ArXiv:
Libraries:
Datasets
WebDataset
License:
OpenDialog / README.md
zhu-han's picture
Update README.md
396cf74 verified
|
raw
history blame
396 Bytes
metadata
license: cc-by-nc-4.0

OpenDialog

OpenDialog is a 6.8k hours spoken dialogue dataset, introduced in paper ZipVoice-Dialog. Check the paper for details

It consists of 1759 hours of Chinese data and 5074 hours of English data.

Data uploading, comming soon...

Full data is already available at https://www.modelscope.cn/datasets/k2-fsa/OpenDialog.