metadata
dataset_info:
features:
- name: audio
dtype: string
- name: raw_transcript
dtype: string
- name: clean_transcript
dtype: string
- name: news
dtype: string
splits:
- name: train
num_bytes: 8253885
num_examples: 506
download_size: 4812714
dataset_size: 8253885
configs:
- config_name: default
data_files:
- split: train
path: data/train-*