Datasets:
metadata
language:
- en
license: mit
size_categories:
- 10K<n<100K
task_categories:
- text-to-speech
- text-to-audio
tags:
- my little pony
- pony preservation project
dataset_info:
features:
- name: speaker
dtype: string
- name: source
dtype: string
- name: start
dtype: float64
- name: end
dtype: float64
- name: style
dtype: string
- name: noise
dtype: string
- name: transcription
dtype: string
- name: audio
dtype: audio
- name: duration
dtype: float64
splits:
- name: train
num_bytes: 15307591666.101
num_examples: 63947
download_size: 14794177233
dataset_size: 15307591666.101
configs:
- config_name: default
data_files:
- split: train
path: data/train-*