metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 25765300382.428
num_examples: 92881
- name: validation
num_bytes: 2914287365.248
num_examples: 10722
- name: test
num_bytes: 558193517.986
num_examples: 2141
download_size: 25249041391
dataset_size: 29237781265.662003
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*