metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 12228343953.8
num_examples: 52100
- name: validation
num_bytes: 1686464237.874
num_examples: 6101
- name: test
num_bytes: 586658790.82
num_examples: 2095
download_size: 15638319440
dataset_size: 14501466982.494
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*