metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcript
dtype: string
splits:
- name: train
num_bytes: 875353186.485
num_examples: 2581
- name: validation
num_bytes: 240146581
num_examples: 651
download_size: 1081812503
dataset_size: 1115499767.4850001
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*