metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 18287257473.824
num_examples: 63166
- name: validation
num_bytes: 2205427112.834
num_examples: 7107
- name: test
num_bytes: 556435471.84
num_examples: 1785
download_size: 21148062014
dataset_size: 21049120058.498
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*