metadata
dataset_info:
features:
- name: Filename
dtype: string
- name: Expression
dtype: string
- name: transcription
dtype: string
- name: output
dtype: string
- name: style_vector
sequence: float64
splits:
- name: train
num_bytes: 13717077.36010252
num_examples: 4605
download_size: 9899158
dataset_size: 13717077.36010252
configs:
- config_name: default
data_files:
- split: train
path: data/train-*