metadata
dataset_info:
features:
- name: source
dtype: string
- name: text
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 24000
splits:
- name: train
num_bytes: 6232193112
num_examples: 11906
download_size: 5724075283
dataset_size: 6232193112
configs:
- config_name: default
data_files:
- split: train
path: data/train-*