metadata
dataset_info:
features:
- name: utt_id
dtype: string
- name: speaker_id
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 762148952
num_examples: 13336
- name: test
num_bytes: 83743725
num_examples: 1481
download_size: 742633085
dataset_size: 845892677
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*