metadata
dataset_info:
features:
- name: id
dtype: string
- name: language
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
splits:
- name: dev
num_bytes: 1886862795.984
num_examples: 23892
- name: train
num_bytes: 12913911474.48
num_examples: 122890
download_size: 14483202989
dataset_size: 14800774270.463999
configs:
- config_name: default
data_files:
- split: dev
path: data/dev-*
- split: train
path: data/train-*