metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: ' transcription'
dtype: string
splits:
- name: train
num_bytes: 4673979585.504
num_examples: 1648
download_size: 4244729405
dataset_size: 4673979585.504