add faq and update live2d

docs/faq.md

@@ -0,0 +1,79 @@
+---
+sidebar_position: 6
+---
+
+# 常见问题
+
+## 部署相关
+
+### 如何选择合适的 LLM?
+- 如果你希望快速部署且不想下载模型,建议使用 OpenAI Compatible API 或 Groq API
+- 如果你想使用本地部署,建议根据显存大小选择合适的模型
+- 详细说明请参考[LLM 配置指南](./user-guide/backend/llm.md)
+
+### 遇到代理相关问题怎么办?
+- 如果你在中国大陆,建议开启代理下载资源
+- 如果本地服务无法访问,需要设置代理绕过本地地址
+- 详细说明请参考[快速开始](./quick-start.md)
+
+### 遇到 "Error calling the chat endpoint..." 错误怎么办?
+常见原因和解决方案:
+- 检查 Ollama 服务是否正常运行
+- 确认模型名称是否与 `ollama list` 列出的完全一致
+- 详细说明请参考[快速开始](./quick-start.md)
+
+### uv 安装后无法找到
+- 如果你是用 `winget` 或者 `curl` 安装的 uv，需要重启命令行或者重新加载 shell 配置文件才能使环境变量生效。
+- 详细说明请参考[快速开始](./quick-start.md)
+
+## 客户端相关
+
+### Web 显示 `{"detail": "Not Found"}` 怎么办
+- 可能是因为没有 `git submodule update --init --recursive` 或者这个命令之前的运行失败了。
+
+### 使用远程的 Live2D 模型链接，报错 `Failed to load LiveD model: Error: Network error` 怎么办
+- 这通常是因为 Web 使用 HTTP 协议无法加载 HTTPS 资源导致的
+- 可以在浏览器设置中允许网站加载不安全内容(Insecure content)来解决:
+  - Chrome: 点击地址栏右侧的盾牌图标 -> 网站设置 -> 不安全内容 -> 允许
+  - Firefox: 点击地址栏左侧的锁图标 -> 关闭连接保护
+  - Edge: 点击地址栏右侧的锁图标 -> 网站权限 -> 不安全内容 -> 允许
+
+### 麦克风无法使用怎么办
+- 请确保已授予浏览器或应用程序麦克风使用权限
+- 检查麦克风输入音量是否合适 - 声音过小或过短可能无法触发语音检测。你可以在设置中调整检测阈值，详细设置请参考 [Web 模式](./user-guide/frontend/web.md)
+- 如果你使用的是 Electron 桌面应用，尝试重启应用。
+- 如果以上方法都无效，可以访问 https://www.vad.ricky0123.com/ 测试麦克风功能。如果该网站也无法正常使用麦克风，可能是系统音频设置或硬件问题
+
+### 桌面客户端无法安装(Windows 已保护你的电脑)或者提示"已损坏"怎么办
+- Windows 用户可以点击"更多信息"然后选择"仍要运行"
+- macOS 用户需要调整系统设置并执行特定命令
+- 详细解决方案请参考[模式介绍](./user-guide/frontend/mode.md)
+
+### 远程访问 Web 界面时麦克风/摄像头/录屏无法使用怎么办?
+- Electron 应用暂时不支持录屏 (`Failed to start screen capture:
+NotSupportedError: Not supported`)。
+- 这是因为这些功能需要安全上下文(HTTPS 或 localhost)
+- 如需远程使用,必须为 Web 服务器配置 HTTPS
+- 详细说明请参考[模式介绍](./user-guide/frontend/mode.md)
+
+### iOS 设备上语音功能异常怎么办?
+- 这是由于 iOS 的安全限制导致的已知问题
+- 需要不断点击才能使说话功能正常运行
+- 详细说明请参考[模式介绍](./user-guide/frontend/mode.md)
+
+### Electron 应用 切换模式导致屏幕一片空白怎么办
+- 属于小概率难以修复的 bug，实测几乎不会触发。
+- 等待几秒，或者重启应用即可解决。
+
+## 其他问题
+
+### 如何获取前端最新版本?
+- 目前不支持自动更新
+- 可以关注 Github Release 页面获取最新版本
+- 详细说明请参考[模式介绍](./user-guide/frontend/mode.md)
+
+### 项目支持哪些使用模式?
+- Web 模式:通过浏览器访问
+- 窗口模式:桌面客户端的默认模式
+- 桌宠模式:提供背景透明、全局置顶的桌面伴侣体验
+- 详细说明请参考[模式介绍](./user-guide/frontend/mode.md)

docs/quick-start.md

@@ -327,7 +327,7 @@ ollama list
 
 ### 4. 配置其他模块
 
-本项目 `conf.yaml` 默认配置中使用了 sherpa-onnx-asr (SenseVoiceSmall) 和 edgeTTS（需要代理），并默认关闭翻译功能，你可以不用进行修改。
+本项目 `conf.yaml` 默认配置中使用了 sherpa-onnx-asr (SenseVoiceSmall) 和 edgeTTS，并默认关闭翻译功能，你可以不用进行修改。
 
 或者你可以参考 [ASR 配置指南](./user-guide/backend/asr.md)、[TTS 配置指南](./user-guide/backend/tts.md) 和 [Translator 配置指南](./user-guide/backend/translate.md) 进行修改。
 

docs/user-guide/_category_.json

@@ -5,5 +5,4 @@
     "type": "generated-index",
     "slug": "/user-guide"
   }
-
 }

docs/user-guide/live2d.md

@@ -110,6 +110,10 @@ Live2D 模型的配置文件是一个包含表情、动作等多个设置项的
 - 支持本地路径和远程 URL
 - 本地路径以 `/live2d-models/` 开头，而非 `./live2d-models/` 开头。
 - 远程 URL 需要指向有效的 `.model.json` 或 `.model3.json` 文件，比如 `https://cdn.jsdelivr.net/gh/guansss/pixi-live2d-display/test/assets/shizuku/shizuku.model.json`
+- 如果使用 HTTP 协议访问 Web，加载 HTTPS 资源时可能会报错 `Failed to load LiveD model: Error: Network error`。此时需要在浏览器设置中允许网站加载不安全内容:
+  - Chrome: 点击地址栏右侧的盾牌图标 -> 网站设置 -> 不安全内容 -> 允许
+  - Firefox: 点击地址栏左侧的锁图标 -> 关闭连接保护
+  - Edge: 点击地址栏右侧的锁图标 -> 网站权限 -> 不安全内容 -> 允许
 
 ### 4.2 显示配置
 

i18n/en/docusaurus-plugin-content-docs/current.json

@@ -62,5 +62,9 @@
   "sidebar.docSidebar.category.社区与贡献.link.generated-index.description": {
     "message": "Developer documentation for Open LLM Vtuber",
     "description": "The generated-index page description for category Community and Contribution in sidebar docSidebar"
+  },
+  "sidebar.docSidebar.category.常见问题": {
+    "message": "FAQ",
+    "description": "The label for category FAQ in sidebar docSidebar"
   }
 }

i18n/en/docusaurus-plugin-content-docs/current/faq.md

@@ -0,0 +1,79 @@
+---
+sidebar_position: 6
+---
+
+# FAQ
+
+## Deployment Related
+
+### How to choose a suitable LLM?
+- If you want quick deployment without downloading models, OpenAI Compatible API or Groq API is recommended
+- For local deployment, choose a model based on your available GPU memory
+- For detailed information, please refer to [LLM Configuration Guide](./user-guide/backend/llm.md)
+
+### What to do with proxy-related issues?
+- If you're in mainland China, it's recommended to use a proxy for downloading resources
+- If local services are inaccessible, you need to configure proxy settings to bypass local addresses
+- For detailed information, please refer to [Quick Start](./quick-start.md)
+
+### What to do when encountering "Error calling the chat endpoint..."?
+Common causes and solutions:
+- Check if the Ollama service is running properly
+- Verify if the model name matches exactly with what's listed in `ollama list`
+- For detailed information, please refer to [Quick Start](./quick-start.md)
+
+### uv not found after installation
+- If you installed uv using `winget` or `curl`, you need to restart your terminal or reload your shell configuration file for the environment variables to take effect.
+- For detailed information, please refer to [Quick Start](./quick-start.md)
+
+## Client Related
+
+### What to do when Web shows `{"detail": "Not Found"}`?
+- This might happen if you haven't run `git submodule update --init --recursive` or if the command failed previously.
+
+### What to do when using remote LiveD model links and getting `Failed to load LiveD model: Error: Network error`?
+- This usually occurs when the webpage using HTTP protocol cannot load HTTPS resources
+- You can resolve this by allowing insecure content in your browser settings:
+  - Chrome: Click the shield icon in the address bar -> Site Settings -> Insecure content -> Allow
+  - Firefox: Click the lock icon in the address bar -> Turn off Enhanced Tracking Protection
+  - Edge: Click the lock icon in the address bar -> Site permissions -> Insecure content -> Allow
+
+### What to do when the microphone isn't working?
+- Make sure you've granted microphone permissions to your browser or application
+- Check if the microphone input volume is appropriate - too low or too short sounds might not trigger voice detection. You can adjust detection thresholds in settings, see [Web Mode](./user-guide/frontend/web.md) for details
+- If you're using the Electron desktop app, try restarting the application
+- If none of the above works, visit https://www.vad.ricky0123.com/ to test microphone functionality. If the site also can't use the microphone, it might be a system audio setting or hardware issue
+
+### What to do when desktop client won't install (Windows protected your PC) or shows "damaged" message?
+- Windows users can click "More info" and then "Run anyway"
+- macOS users need to adjust system settings and execute specific commands
+- For detailed solutions, please refer to [Mode Introduction](./user-guide/frontend/mode.md)
+
+### What to do when microphone/camera/screen capture doesn't work with remote Web interface access?
+- Electron app currently doesn't support screen capture (`Failed to start screen capture:
+NotSupportedError: Not supported`)
+- These features require a secure context (HTTPS or localhost)
+- For remote access, you must configure HTTPS for the web server
+- For detailed information, please refer to [Mode Introduction](./user-guide/frontend/mode.md)
+
+### What to do about voice functionality issues on iOS devices?
+- This is a known issue due to iOS security restrictions
+- Continuous tapping is required for the speech function to work properly
+- For detailed information, please refer to [Mode Introduction](./user-guide/frontend/mode.md)
+
+### What to do when Electron app shows blank screen after switching modes?
+- This is a rare bug that's difficult to fix, but it rarely occurs in practice
+- Wait a few seconds or restart the application to resolve the issue
+
+## Other Questions
+
+### How to get the latest frontend version?
+- Automatic updates are currently not supported
+- You can follow the Github Release page for latest versions
+- For detailed information, please refer to [Mode Introduction](./user-guide/frontend/mode.md)
+
+### What usage modes does the project support?
+- Web Mode: Access through browser
+- Window Mode: Default mode for desktop client
+- Desktop Pet Mode: Provides transparent background and global top-most desktop companion experience
+- For detailed information, please refer to [Mode Introduction](./user-guide/frontend/mode.md)

i18n/en/docusaurus-plugin-content-docs/current/quick-start.md

@@ -321,7 +321,7 @@ For detailed explanations of the configuration file, refer to [User Guide/Config
 
 ### 4. Configure Other Modules
 
-The default configuration in this project's `conf.yaml` uses sherpa-onnx-asr (SenseVoiceSmall) and edgeTTS (requires a proxy), and translation is disabled by default, so you do not need to make modifications.
+The default configuration in this project's `conf.yaml` uses sherpa-onnx-asr (SenseVoiceSmall) and edgeTTS, and translation is disabled by default, so you do not need to make modifications.
 
 Alternatively, you can refer to the [ASR Configuration Guide](./user-guide/backend/asr.md), [TTS Configuration Guide](./user-guide/backend/tts.md), and [Translator Configuration Guide](./user-guide/backend/translate.md) for modifications.
 

i18n/en/docusaurus-plugin-content-docs/current/user-guide/live2d.md

@@ -110,6 +110,11 @@ Add the model configuration in the `model_dict.json` file in the project root di
 - Local paths should start with `/live2d-models/`, not `./live2d-models/`
 - Remote URLs should point to valid `.model.json` or `.model3.json` files, e.g., `https://cdn.jsdelivr.net/gh/guansss/pixi-live2d-display/test/assets/shizuku/shizuku.model.json`
 
+- When loading HTTPS resources while accessing the Web via HTTP protocol, you may encounter the error `Failed to load LiveD model: Error: Network error`. In this case, you need to allow the website to load insecure content in your browser settings:
+  - Chrome: Click the shield icon on the right side of the address bar -> Site Settings -> Insecure content -> Allow
+  - Firefox: Click the lock icon on the left side of the address bar -> Turn off Enhanced Tracking Protection
+  - Edge: Click the lock icon on the right side of the address bar -> Site permissions -> Insecure content -> Allow
+
 ### 4.2 Display Configuration
 
 | Config Item     | Description                      | Recommended Value |