metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 24000
- name: text
dtype: string
splits:
- name: train
num_bytes: 4991145359
num_examples: 13100
download_size: 4946993107
dataset_size: 4991145359
configs:
- config_name: default
data_files:
- split: train
path: data/train-*