metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sentence
dtype: string
splits:
- name: train
num_bytes: 6571743863
num_examples: 10000
download_size: 5452936696
dataset_size: 6571743863
configs:
- config_name: default
data_files:
- split: train
path: data/train-*