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: 10670017826.486773
num_examples: 92965
- name: validation
num_bytes: 31167039.82322723
num_examples: 279
download_size: 10680316305
dataset_size: 10701184866.31
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*