metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 12514005492.18
num_examples: 78030
- name: validation
num_bytes: 2031649680.85
num_examples: 7809
- name: test
num_bytes: 571419545.465
num_examples: 1977
download_size: 23511394867
dataset_size: 15117074718.495
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*