metadata
dataset_info:
features:
- name: audio_id
dtype: string
- name: transcription
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 7723539252.875
num_examples: 52081
download_size: 7705893691
dataset_size: 7723539252.875
configs:
- config_name: default
data_files:
- split: train
path: data/train-*