metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 23229945394.616
num_examples: 76628
- name: validation
num_bytes: 2891725263.952
num_examples: 9076
- name: test
num_bytes: 589607847.4
num_examples: 1930
download_size: 24492482702
dataset_size: 26711278505.968002
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*