metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 29731215136.349655
num_examples: 206184
- name: test
num_bytes: 301253908.97834307
num_examples: 2083
download_size: 29791478459
dataset_size: 30032469045.328
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*