metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sentence
dtype: string
- name: source
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 6487285193.635
num_examples: 26035
- name: test
num_bytes: 1641288622.114
num_examples: 6509
download_size: 7348735079
dataset_size: 8128573815.749001
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*