metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcripts
dtype: string
splits:
- name: train
num_bytes: 27336052039.712
num_examples: 358537
download_size: 24519982590
dataset_size: 27336052039.712
configs:
- config_name: default
data_files:
- split: train
path: data/train-*