metadata
dataset_info:
features:
- name: line_id
dtype: string
- name: audio
dtype: audio
- name: text
dtype: string
- name: speaker_id
dtype: int64
splits:
- name: train
num_bytes: 113660513
num_examples: 125
download_size: 104276275
dataset_size: 113660513
configs:
- config_name: default
data_files:
- split: train
path: data/train-*