Datasets:
Tasks:
Audio Classification
Modalities:
Audio
Formats:
arrow
Languages:
Russian
Size:
10K - 100K
Tags:
WakeWord
Download dataset_info.json from MiGo04/WakeWordDetection: direct link, hf CLI and curl.
- Browser
- Download file 224 Bytes
-
https://huggingface.co/datasets/MiGo04/WakeWordDetection/resolve/main/dataset_info.json
- Command line
-
hf download hf://datasets/MiGo04/WakeWordDetection/dataset_info.json
-
curl -L -o dataset_info.json https://huggingface.co/datasets/MiGo04/WakeWordDetection/resolve/main/dataset_info.json
224 Bytes
| { | |
| "citation": "", | |
| "description": "", | |
| "features": { | |
| "audio": { | |
| "_type": "Audio" | |
| }, | |
| "label": { | |
| "dtype": "int64", | |
| "_type": "Value" | |
| } | |
| }, | |
| "homepage": "", | |
| "license": "" | |
| } |