metadata
dataset_info:
features:
- name: original_text
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
splits:
- name: train
num_bytes: 32237597915.917706
num_examples: 202871
- name: validation
num_bytes: 43704937.554294854
num_examples: 279
download_size: 29723074855
dataset_size: 32281302853.472
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*