metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: name
dtype: string
- name: time
dtype: float64
- name: size
dtype: int64
- name: text
dtype: string
- name: hara
dtype: string
splits:
- name: train
num_bytes: 28374237968.048
num_examples: 31184
download_size: 15873913436
dataset_size: 28374237968.048
configs:
- config_name: default
data_files:
- split: train
path: data/train-*