metadata
dataset_info:
features:
- name: key
dtype: int64
- name: audio
dtype: audio
- name: transcript
dtype: string
- name: language
dtype: string
splits:
- name: train
num_bytes: 3797549573
num_examples: 3596
- name: validation
num_bytes: 936458758
num_examples: 898
download_size: 4527614254
dataset_size: 4734008331
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*