Hugging Face's logo

Datasets:
snapwre
/
amharic-speech

Tasks:
Automatic Speech Recognition
Modalities:
Audio
Text
Formats:
parquet
Languages:
Amharic
Size:
10K - 100K
Tags:
speech
ethiopia
low-resource
crowdsourced
Libraries:
Datasets
Dask
Polars
License:
Dataset card Data Studio Files
xet
 Community
1
amharic-speech / data

Commit History

  1. Remove shards superseded by this release

    Chapimenge
    Verified signature
    4861d64

  2. Add files using upload-large-folder tool

    Chapimenge
    Verified signature
    8c6d529

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

    Chapimenge
    Verified signature
    d35c0a6