metadata
dataset_info:
features:
- name: id
dtype: string
- name: language
dtype: string
- name: score
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
splits:
- name: train
num_bytes: 142846713396.732
num_examples: 3227
download_size: 125597399780
dataset_size: 142846713396.732
configs:
- config_name: default
data_files:
- split: train
path: data/train-*