metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcript
dtype: string
- name: duration
dtype: float64
- name: filename
dtype: string
splits:
- name: train
num_bytes: 30215675
num_examples: 200
download_size: 28905548
dataset_size: 30215675
configs:
- config_name: default
data_files:
- split: train
path: data/train-*