metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 19420811687.516
num_examples: 87227
- name: validation
num_bytes: 2256067886.578
num_examples: 10039
- name: test
num_bytes: 476489105.684
num_examples: 2058
download_size: 25969082991
dataset_size: 22153368679.778
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*