metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 44100
- name: text
dtype: string
- name: filename
dtype: string
- name: language
dtype: string
splits:
- name: train
num_bytes: 3729918060.526
num_examples: 5526
- name: validation
num_bytes: 370862854
num_examples: 545
download_size: 3564461911
dataset_size: 4100780914.526
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*