metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 44100
- name: text
dtype: string
- name: speaker_id
dtype: string
- name: style
dtype: string
- name: id
dtype: string
splits:
- name: train
num_bytes: 813330192.875
num_examples: 2905
download_size: 783497413
dataset_size: 813330192.875
configs:
- config_name: default
data_files:
- split: train
path: data/train-*