metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
- name: speaker_id
dtype: int64
splits:
- name: train
num_bytes: 6900898092.656
num_examples: 12852
download_size: 6379993122
dataset_size: 6900898092.656
configs:
- config_name: default
data_files:
- split: train
path: data/train-*