metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 3536214502.768
num_examples: 14776
- name: validation
num_bytes: 569844031.584
num_examples: 2132
- name: test
num_bytes: 545707152.356
num_examples: 2059
download_size: 4573384472
dataset_size: 4651765686.708
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*