metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 1918145060.272
num_examples: 27702
- name: test
num_bytes: 835022060.538
num_examples: 11873
download_size: 2822876508
dataset_size: 2753167120.81
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*