metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcripts
dtype: string
splits:
- name: train
num_bytes: 2950680921.381188
num_examples: 6366
- name: test
num_bytes: 759706762.1288122
num_examples: 1592
download_size: 3292904979
dataset_size: 3710387683.51
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*