metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 8271807364.2
num_examples: 46755
- name: validation
num_bytes: 970254510.506
num_examples: 5109
- name: test
num_bytes: 521374082.878
num_examples: 2719
download_size: 9744499883
dataset_size: 9763435957.584
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*