metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 9207964781.4
num_examples: 26232
- name: validation
num_bytes: 938019733.352
num_examples: 3082
- name: test
num_bytes: 590044628.818
num_examples: 1919
download_size: 8676891681
dataset_size: 10736029143.57
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*