metadata
dataset_info:
features:
- name: text
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 380771626147.464
num_examples: 2065541
download_size: 378684038278
dataset_size: 380771626147.464
configs:
- config_name: default
data_files:
- split: train
path: data/train-*