metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 6583248225.582
num_examples: 22207
- name: validation
num_bytes: 809739893.892
num_examples: 2763
- name: test
num_bytes: 573054235.814
num_examples: 1957
download_size: 7417729189
dataset_size: 7966042355.288
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*