metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: transcription
dtype: string
- name: audio
dtype: audio
- name: text_description
dtype: string
splits:
- name: train
num_bytes: 56962974
num_examples: 20
download_size: 52086217
dataset_size: 56962974
configs:
- config_name: default
data_files:
- split: train
path: data/train-*