metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: preferred_text
dtype: string
splits:
- name: test
num_bytes: 168222354.49
num_examples: 2927
- name: train
num_bytes: 689638875.5
num_examples: 11868
download_size: 2143483418
dataset_size: 857861229.99
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
- split: train
path: data/train-*