metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: id
dtype: string
splits:
- name: train
num_bytes: 2954600348.4302382
num_examples: 22152
- name: test
num_bytes: 66689245.85658718
num_examples: 500
download_size: 2982696792
dataset_size: 3021289594.2868257