Tingshu (听舒)

让作者的声音直达听众 | Bringing the author's voice directly to you

项目预览 | Preview

点击进入项目主页｜Home

快速开始 | Quick Start

点击这里观看视频

1. 安装环境 | Install Environment

./scripts/install.sh

2. 下载模型权重 | Download Model Weights

./scripts/download_weights.sh

3. 启动 Web UI | Run Web Interface

./scripts/run_gradio.sh

4. 生成前端音频数据 | Generate Audio Data

./scripts/generate_web_data.sh

项目优化 ｜ TODO

• 1. 使用新方案替代tts方式生成的录音缺乏情感，顿挫，听感不好问题。
  Replace the TTS-generated recordings to address issues of lacking emotion, intonation, and poor listening experience.

• 2. 更新web ui, 替换静态加载模式
  Update web ui and switch from static loading mode.

• 3. 增加英文版自传数据内容
  Add autobiographical content in the English version.

• 4. 适当增加背景音乐内容、音效
  Appropriately enhance background music and sound effects.

免责声明 | Disclaimer

本项目仅供学习和研究使用。使用者须遵守当地的法律法规，包括但不限于 DMCA 相关法律。我们不对任何非法使用承担责任。

This project is for research and learning purposes only. Users must comply with local laws and regulations, including but not limited to DMCA-related laws. We do not take any responsibility for illegal usage.

技术鸣谢 | Credits

本项目基于以下开源项目构建：

• VITS2
• Bert-VITS2
• GPT VITS
• GPT-SoVITS
• Fish Speech