metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 19535432121.256
num_examples: 76628
- name: validation
num_bytes: 2829912222.048
num_examples: 9076
- name: test
num_bytes: 590831868.84
num_examples: 1930
download_size: 24440529990
dataset_size: 22956176212.144
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*