metadata
dataset_info:
features:
- name: text
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 24000
splits:
- name: train
num_bytes: 3050475454.561443
num_examples: 4768
download_size: 3566230681
dataset_size: 3050475454.561443
configs:
- config_name: default
data_files:
- split: train
path: data/train-*