birgermoell/synthetic_compassion

---
dataset_info:
features:
- name: audio
dtype: audio
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 117098145
num_examples: 439
download_size: 111238398
dataset_size: 117098145
license: mit
language:
- en
pretty_name: Synthetic Compassion
size_categories:
- n<1K
---
# Dataset Card for "synthetic_compassion"
Synthetic compassion is a dataset consisting of generated compassionate speech for meditation and psychology advice.
The dataset consists of unique voices as each audio file has its own synthetic voice.
The speech is generated with Tortoise TTS

## How to use
Here is how you can load the dataset
```python
from datasets import load_dataset
dataset = load_dataset("birgermoell/synthetic_compassion")
```

原始信息汇总

数据集概述

基本信息

• 数据集名称: Synthetic Compassion
• 数据集大小: 117098145 字节
• 下载大小: 111238398 字节
• 许可证: MIT
• 语言: 英语
• 规模类别: 小于1K

数据集内容

• 特征:
  • audio: 音频数据
  • transcription: 文本数据

数据划分

• 训练集:
  • 示例数量: 439
  • 数据大小: 117098145 字节