metadata
dataset_info:
features:
- name: transcription
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 1184495678.0013225
num_examples: 25741
- name: test
num_bytes: 62435958.81567758
num_examples: 1355
download_size: 1249090411
dataset_size: 1246931636.8170002
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*