metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 34710823186.014
num_examples: 144021
- name: validation
num_bytes: 3752482198.924
num_examples: 15378
- name: test
num_bytes: 529945896.086
num_examples: 1922
download_size: 44212191018
dataset_size: 38993251281.024
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*