metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: language
dtype: string
splits:
- name: train
num_bytes: 895194553
num_examples: 55502
- name: validation
num_bytes: 111517323
num_examples: 6938
- name: test
num_bytes: 112475729
num_examples: 6938
download_size: 1109401806
dataset_size: 1119187605
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*