Hugging Face's logo

Datasets:
google
/
WaxalNLP

Tasks:
Automatic Speech Recognition
Text-to-Speech
Modalities:
Audio
Text
Formats:
parquet
Languages:
Acoli
Akan
Amharic
Size:
1M - 10M
ArXiv:
Tags:
audio
automatic-speech-recognition
text-to-speech
Libraries:
Datasets
Dask
Polars
License:
Dataset card Data Studio Files
xet
 Community
26
WaxalNLP / data /ASR_v2 /metadata /split_map
14.7 MB
  • 2 contributors
 History: 1 commit
perrynelson's picture
perrynelson
Add ASR v2 speaker-disjoint split definitions
96ced60