---
title: dania
canonical_url: "https://www.modelscope.cn/models/villia/dania"
md_url: "https://www.modelscope.cn/models/villia/dania.md"
repository: villia/dania
chinese_name: "达妮娅 GPT-SoVITS"
last_updated: 2026-05-25
license: "Apache License 2.0"
pipeline_tag: text-to-speech
tasks:
  - text-to-speech
library_name:
  - pytorch
frameworks:
  - PyTorch
language:
  - zh
downloads: 111
stars: 11
---

# dania

> dania - villia 在 ModelScope 开源的模型。达妮娅GPT-Sovits模型

villia/dania 是 ModelScope 魔搭社区上的text-to-speech模型，采用 Apache License 2.0 许可。

- **Repository**: villia/dania
- **License**: Apache License 2.0
- **Tasks**: text-to-speech
- **Downloads**: 111
- **Stars**: 11
- **Last updated**: 2026-05-25

Source: https://www.modelscope.cn/models/villia/dania

---

# 达妮娅GPT-SoVITS模型
gpt-sovits整合包地址    
https://www.yuque.com/baicaigongchang1145haoyuangong/ib3g1e/dkxgpiy9zb96hob4

## 推荐参考音频
output/slicer_opt/output.wav_0009342720_0009558400.wav      
怎么啊？如果有你在也不放心，那就干脆给我也装个限制器或者炸弹喽。     

确保把参考音频【output.wav_0009342720_0009558400.wav】放到了output\slicer_opt目录下      
提供的各文件对应位置      
GPT_SoVITS/dania-e15.ckpt    
GPT_SoVITS/dania_e16_s2192.pth    
GPT_weights_v2ProPlus/dania-e15.ckpt    
SoVITS_weights_v2ProPlus/dania_e16_s2192.pth    
GPT_SoVITS/configs/tts_infer.yaml    
api模式.bat复制到根目录即可    

怎么还要我写200字才给通过。

# python代码示例
```python
import requests

url = "http://127.0.0.1:9880/tts"

params = {
    "text": "拉海洛来了个救世主，人们说攻略任何女生对他来说只是时间问题，但对于5月21日的达妮娅，时间却成了最大的问题",
    "text_lang": "zh",
    "ref_audio_path": "output/slicer_opt/output.wav_0009342720_0009558400.wav",
    "prompt_text": "怎么啊？如果有你在也不放心，那就干脆给我也装个限制器或者炸弹喽。",
    "prompt_lang": "zh",
}

response = requests.get(url, params=params)

if response.status_code == 200:
    with open("output.wav", "wb") as f:
        f.write(response.content)
    print("保存成功：output.wav")
else:
    print("失败：", response.json())
```