metadata
dataset_info:
features:
- name: transcription
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 1474007827
num_examples: 96554
download_size: 804197923
dataset_size: 1474007827
configs:
- config_name: default
data_files:
- split: train
path: data/train-*