metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
splits:
- name: train
num_bytes: 1992738327
num_examples: 3358
download_size: 1509612493
dataset_size: 1992738327
configs:
- config_name: default
data_files:
- split: train
path: data/train-*