metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 9832223980.228
num_examples: 43283
- name: validation
num_bytes: 1058019174.2
num_examples: 4856
- name: test
num_bytes: 511010243.504
num_examples: 2476
download_size: 9947826187
dataset_size: 11401253397.932001
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*