Datasets:

Blgn94
/

mongolian-stt-dataset

Tasks:

Automatic Speech Recognition

Modalities:

Formats:

Languages:

Size:

Libraries:

License:

Dataset card Data Studio Files Files and versions

mongolian-stt-dataset

5.21 GB

1 contributor

History: 5 commits

Blgn94's picture

Update card: v24 corpus, YouTube removed, honest eval caveats

d10e526 verified 4 days ago

data
Replace v20 corpus with v24; remove all YouTube-sourced audio (598 clips, ~1.1h). Now 146.9h read speech only. 4 days ago
.gitattributes

2.5 kB
initial commit 17 days ago
README.md

4.51 kB
Update card: v24 corpus, YouTube removed, honest eval caveats 4 days ago