Hugging Face's logo

Datasets:
snapwre
/
amharic-speech

Tasks:
Automatic Speech Recognition
Modalities:
Audio
Text
Formats:
parquet
Languages:
Amharic
Size:
1K - 10K
Tags:
speech
ethiopia
low-resource
crowdsourced
Libraries:
Datasets
Dask
Polars
License:
Dataset card Data Studio Files
xet
 Community
1
amharic-speech / metadata.csv

Commit History

Dataset.ET Amharic Speech v0.1.0 — 22.7h, 7,405 clips, 320 speakers
d35c0a6
verified

Chapimenge commited on