Datasets:
Tasks:
Automatic Speech Recognition
Formats:
parquet
Languages:
English
Size:
1M - 10M
ArXiv:
License:
Download validation/dev.json from MLCommons/peoples_speech: direct link, hf CLI and curl.
- Browser
- Download file 4.27 MB
-
https://huggingface.co/datasets/MLCommons/peoples_speech/resolve/refs%2Fpr%2F2/validation/dev.json
- Command line
-
hf download hf://datasets/MLCommons/peoples_speech@refs/pr/2/validation/dev.json
-
curl -L -o dev.json https://huggingface.co/datasets/MLCommons/peoples_speech/resolve/refs%2Fpr%2F2/validation/dev.json
4.27 MB
File too large to display, you can check the raw version instead.