metadata
dataset_info:
features:
- name: speaker_id
dtype: string
- name: audio
list: float32
- name: text
dtype: string
splits:
- name: train
num_bytes: 159819201
num_examples: 500
download_size: 148637252
dataset_size: 159819201
configs:
- config_name: default
data_files:
- split: train
path: data/train-*