metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 31210995872.924
num_examples: 102566
- name: validation
num_bytes: 3410621920.936
num_examples: 11232
- name: test
num_bytes: 568692146.85
num_examples: 1917
download_size: 32472722938
dataset_size: 35190309940.71
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*