metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sentence
dtype: string
- name: split
dtype: string
splits:
- name: train
num_bytes: 26465079531.408
num_examples: 963636
download_size: 25148856036
dataset_size: 26465079531.408
configs:
- config_name: default
data_files:
- split: train
path: data/train-*