metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 10028021683.128
num_examples: 58096
- name: validation
num_bytes: 1138112066.432
num_examples: 6331
- name: test
num_bytes: 425642509.524
num_examples: 2131
download_size: 16342601901
dataset_size: 11591776259.084
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*