metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 807889708.0629112
num_examples: 24921
- name: test
num_bytes: 32503643.295088924
num_examples: 1000
download_size: 819824750
dataset_size: 840393351.358