Merge pull request #461 from harry0703/dev

Optimize memory usage in moviepy

.github/workflows/gh-pages.yml

@@ -0,0 +1,33 @@
+name: gp-pages
+on:
+  push:
+    branches:
+      - main
+    paths:
+      - 'sites/**'
+
+jobs:
+  build-deploy:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v2
+      - name: Set-up Node
+        uses: actions/setup-node@v1
+        with:
+          node-version: "18.15.0"
+      - name: Install pnpm
+        run: npm install -g pnpm
+      - name: Install dependencies
+        run: pnpm i
+        working-directory: ./sites
+      - name: Build gh-pages
+        run: pnpm docs:build
+        working-directory: ./sites
+      - name: Deploy to gh-pages
+        uses: crazy-max/ghaction-github-pages@v1
+        with:
+          target_branch: gh-pages
+          build_dir: sites/docs/.vuepress/dist
+        env:
+          GITHUB_TOKEN: ${{ secrets.MY_TOKEN }}

.gitignore

@@ -11,4 +11,15 @@
 .vscode
 /**/.streamlit
 __pycache__
-logs/
+logs/
+
+node_modules
+# VuePress 默认临时文件目录
+/sites/docs/.vuepress/.temp
+# VuePress 默认缓存目录
+/sites/docs/.vuepress/.cache
+# VuePress 默认构建生成的静态文件目录
+/sites/docs/.vuepress/dist
+# 模型目录
+/models/
+./models/*

CHANGELOG.md

@@ -0,0 +1,22 @@
+<!-- insertion marker -->
+## [1.1.2](https://github.com/KevinZhang19870314/MoneyPrinterTurbo/releases/tag/1.1.2) - 2024-04-18
+
+<small>[Compare with first commit](https://github.com/KevinZhang19870314/MoneyPrinterTurbo/compare/d4f7b53b841e65da658e3d77822f9923286ddab6...1.1.2)</small>
+
+### Features
+
+- add support for maximum concurrency of /api/v1/videos ([abe12ab](https://github.com/KevinZhang19870314/MoneyPrinterTurbo/commit/abe12abd7b78997651468ad5dd656985066f8bd9) by kevin.zhang).
+- add task deletion endpoint ([d57434e](https://github.com/KevinZhang19870314/MoneyPrinterTurbo/commit/d57434e0d31c8195dbcd3c86ff2763af96736cdf) by kevin.zhang).
+- add redis support for task state management ([3d45348](https://github.com/KevinZhang19870314/MoneyPrinterTurbo/commit/3d453486627234937c7bfe6f176890360074696b) by kevin.zhang).
+- enable cors to allow play video through mounted videos url ([3b1871d](https://github.com/KevinZhang19870314/MoneyPrinterTurbo/commit/3b1871d591873594bb4aa8dc17a1253b3a7563a3) by kevin.zhang).
+- add /api/v1/get_bgm_list and /api/v1/upload_bgm_file ([6d8911f](https://github.com/KevinZhang19870314/MoneyPrinterTurbo/commit/6d8911f5bf496e7c5dd718309a302df88d11817b) by cathy).
+- return combined videos in /api/v1/tasks response ([28199c9](https://github.com/KevinZhang19870314/MoneyPrinterTurbo/commit/28199c93b78f67e9a6bf50f290f1591078f63da8) by cathy).
+- add Dockerfile ([f3b3c7f](https://github.com/KevinZhang19870314/MoneyPrinterTurbo/commit/f3b3c7fb47b01ed4ecba44eaebf29f5d6d2cb7b5) by kevin.zhang).
+
+### Bug Fixes
+
+- response parsing bug for gemini ([ee7306d](https://github.com/KevinZhang19870314/MoneyPrinterTurbo/commit/ee7306d216ea41e40855bbca396cacb094d572db) by elf-mouse).
+
+### Code Refactoring
+
+- Streaming MP4 files in the browser using video html element instead of waiting for the entire file to download before playing ([d13a3cf](https://github.com/KevinZhang19870314/MoneyPrinterTurbo/commit/d13a3cf6e911d1573c62b1f6459c3c0b7a1bc18d) by kevin.zhang).

Dockerfile

@@ -1,9 +1,12 @@
 # Use an official Python runtime as a parent image
-FROM python:3.10-slim
+FROM python:3.10-slim-bullseye
 
 # Set the working directory in the container
 WORKDIR /MoneyPrinterTurbo
 
+# 设置/MoneyPrinterTurbo目录权限为777
+RUN chmod 777 /MoneyPrinterTurbo
+
 ENV PYTHONPATH="/MoneyPrinterTurbo"
 
 # Install system dependencies
@@ -16,12 +19,15 @@ RUN apt-get update && apt-get install -y \
 # Fix security policy for ImageMagick
 RUN sed -i '/<policy domain="path" rights="none" pattern="@\*"/d' /etc/ImageMagick-6/policy.xml
 
-# Copy the current directory contents into the container at /MoneyPrinterTurbo
-COPY . .
+# Copy only the requirements.txt first to leverage Docker cache
+COPY requirements.txt ./
 
 # Install Python dependencies
 RUN pip install --no-cache-dir -r requirements.txt
 
+# Now copy the rest of the codebase into the image
+COPY . .
+
 # Expose the port the app runs on
 EXPOSE 8501
 

README-en.md

@@ -10,9 +10,9 @@
 
 <h3>English | <a href="README.md">简体中文</a></h3>
 
-
-> Thanks to [RootFTW](https://github.com/Root-FTW) for the translation
-
+<div align="center">
+  <a href="https://trendshift.io/repositories/8731" target="_blank"><img src="https://trendshift.io/api/badge/repositories/8731" alt="harry0703%2FMoneyPrinterTurbo | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a>
+</div>
 
 Simply provide a <b>topic</b> or <b>keyword</b> for a video, and it will automatically generate the video copy, video
 materials, video subtitles, and video background music before synthesizing a high-definition short video.
@@ -59,8 +59,7 @@ https://reccloud.com
 - [x] Supports integration with various models such as **OpenAI**, **moonshot**, **Azure**, **gpt4free**, **one-api**,
   **qianwen**, **Google Gemini**, **Ollama** and more
 
- ❓[How to Use the Free OpenAI GPT-3.5 Model?](https://github.com/harry0703/MoneyPrinterTurbo/blob/main/README-en.md#common-questions-)
-
+❓[How to Use the Free OpenAI GPT-3.5 Model?](https://github.com/harry0703/MoneyPrinterTurbo/blob/main/README-en.md#common-questions-)
 
 ### Future Plans 📅
 
@@ -111,6 +110,11 @@ https://reccloud.com
 </tbody>
 </table>
 
+## System Requirements 📦
+
+- Recommended minimum 4 CPU cores or more, 8G of memory or more, GPU is not required
+- Windows 10 or MacOS 11.0, and their later versions
+
 ## Installation & Deployment 📥
 
 - Try to avoid using **Chinese paths** to prevent unpredictable issues
@@ -245,8 +249,28 @@ You can switch between them by modifying the `subtitle_provider` in the `config.
 It is recommended to use `edge` mode, and switch to `whisper` mode if the quality of the subtitles generated is not
 satisfactory.
 
+> Note:
 > If left blank, it means no subtitles will be generated.
 
+**Download whisper**
+- Please ensure a good internet connectivity
+- `whisper` model can be downloaded from HuggingFace: https://huggingface.co/openai/whisper-large-v3/tree/main
+
+After downloading the model to local machine, copy the whole folder and put it into the following path: `.\MoneyPrinterTurbo\models`
+
+This is what the final path should look like: `.\MoneyPrinterTurbo\models\whisper-large-v3`
+
+```
+MoneyPrinterTurbo  
+  ├─models
+  │   └─whisper-large-v3
+  │          config.json
+  │          model.bin
+  │          preprocessor_config.json
+  │          tokenizer.json
+  │          vocabulary.json
+```
+
 ## Background Music 🎵
 
 Background music for videos is located in the project's `resource/songs` directory.
@@ -261,12 +285,16 @@ own fonts.
 ## Common Questions 🤔
 
 ### ❓How to Use the Free OpenAI GPT-3.5 Model?
-[OpenAI has announced that ChatGPT with 3.5 is now free](https://openai.com/blog/start-using-chatgpt-instantly), and developers have wrapped it into an API for direct usage.
+
+[OpenAI has announced that ChatGPT with 3.5 is now free](https://openai.com/blog/start-using-chatgpt-instantly), and
+developers have wrapped it into an API for direct usage.
 
 **Ensure you have Docker installed and running**. Execute the following command to start the Docker service:
+
 ```shell
 docker run -p 3040:3040 missuo/freegpt35
 ```
+
 Once successfully started, modify the `config.toml` configuration as follows:
 
 - Set `llm_provider` to `openai`
@@ -341,4 +369,4 @@ Click to view the [`LICENSE`](LICENSE) file
 
 ## Star History
 
-[![Star History Chart](https://api.star-history.com/svg?repos=harry0703/MoneyPrinterTurbo&type=Date)](https://star-history.com/#harry0703/MoneyPrinterTurbo&Date)
+[![Star History Chart](https://api.star-history.com/svg?repos=harry0703/MoneyPrinterTurbo&type=Date)](https://star-history.com/#harry0703/MoneyPrinterTurbo&Date)

README.md

@@ -9,6 +9,9 @@
 </p>
 <br>
 <h3>简体中文 | <a href="README-en.md">English</a></h3>
+<div align="center">
+  <a href="https://trendshift.io/repositories/8731" target="_blank"><img src="https://trendshift.io/api/badge/repositories/8731" alt="harry0703%2FMoneyPrinterTurbo | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a>
+</div>
 <br>
 只需提供一个视频 <b>主题</b> 或 <b>关键词</b> ，就可以全自动生成视频文案、视频素材、视频字幕、视频背景音乐，然后合成一个高清的短视频。
 <br>
@@ -26,7 +29,6 @@
 ## 特别感谢 🙏
 
 由于该项目的 **部署** 和 **使用**，对于一些小白用户来说，还是 **有一定的门槛**，在此特别感谢
-
 **录咖（AI智能 多媒体服务平台）** 网站基于该项目，提供的免费`AI视频生成器`服务，可以不用部署，直接在线使用，非常方便。
 
 - 中文版：https://reccloud.cn
@@ -34,6 +36,14 @@
 
 ![](docs/reccloud.cn.jpg)
 
+## 感谢赞助 🙏
+
+感谢佐糖 https://picwish.cn 对该项目的支持和赞助，使得该项目能够持续的更新和维护。
+
+佐糖专注于**图像处理领域**，提供丰富的**图像处理工具**，将复杂操作极致简化，真正实现让图像处理更简单。
+
+![picwish.jpg](docs/picwish.jpg)
+
 ## 功能特性 🎯
 
 - [x] 完整的 **MVC架构**，代码 **结构清晰**，易于维护，支持 `API` 和 `Web界面`
@@ -42,15 +52,15 @@
     - [x] 竖屏 9:16，`1080x1920`
     - [x] 横屏 16:9，`1920x1080`
 - [x] 支持 **批量视频生成**，可以一次生成多个视频，然后选择一个最满意的
-- [x] 支持 **视频片段时长**设置，方便调节素材切换频率
+- [x] 支持 **视频片段时长** 设置，方便调节素材切换频率
 - [x] 支持 **中文** 和 **英文** 视频文案
-- [x] 支持 **多种语音** 合成
+- [x] 支持 **多种语音** 合成，可 **实时试听** 效果
 - [x] 支持 **字幕生成**，可以调整 `字体`、`位置`、`颜色`、`大小`，同时支持`字幕描边`设置
 - [x] 支持 **背景音乐**，随机或者指定音乐文件，可设置`背景音乐音量`
-- [x] 视频素材来源 **高清**，而且 **无版权**
-- [x] 支持 **OpenAI**、**moonshot**、**Azure**、**gpt4free**、**one-api**、**通义千问**、**Google Gemini**、**Ollama** 等多种模型接入
-
- ❓[如何使用免费的 **OpenAI GPT-3.5** 模型?](https://github.com/harry0703/MoneyPrinterTurbo?tab=readme-ov-file#%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98-)
+- [x] 视频素材来源 **高清**，而且 **无版权**，也可以使用自己的 **本地素材**
+- [x] 支持 **OpenAI**、**Moonshot**、**Azure**、**gpt4free**、**one-api**、**通义千问**、**Google Gemini**、**Ollama**、
+  **DeepSeek**、 **文心一言** 等多种模型接入
+    - 中国用户建议使用 **DeepSeek** 或 **Moonshot** 作为大模型提供商（国内可直接访问，不需要VPN。注册就送额度，基本够用）
 
 ### 后期计划 📅
 
@@ -59,15 +69,12 @@
 - [ ] 增加视频转场效果，使其看起来更加的流畅
 - [ ] 增加更多视频素材来源，优化视频素材和文案的匹配度
 - [ ] 增加视频长度选项：短、中、长
-- [ ] 打包成一键启动包（Windows，macOS），方便使用
-- [ ] 增加免费网络代理，让访问OpenAI和素材下载不再受限
-- [ ] 可以使用自己的素材
-- [ ] 朗读声音和背景音乐，提供实时试听
-- [ ] 支持更多的语音合成服务商，比如 OpenAI TTS, Azure TTS
+- [ ] 支持更多的语音合成服务商，比如 OpenAI TTS
 - [ ] 自动上传到YouTube平台
 
 ## 交流讨论 💬
-<img src="docs/wechat-03.jpg" width="300">
+
+<img src="docs/wechat-group.jpg" width="250">
 
 ## 视频演示 📺
 
@@ -77,12 +84,14 @@
 <thead>
 <tr>
 <th align="center"><g-emoji class="g-emoji" alias="arrow_forward">▶️</g-emoji> 《如何增加生活的乐趣》</th>
+<th align="center"><g-emoji class="g-emoji" alias="arrow_forward">▶️</g-emoji> 《金钱的作用》<br>更真实的合成声音</th>
 <th align="center"><g-emoji class="g-emoji" alias="arrow_forward">▶️</g-emoji> 《生命的意义是什么》</th>
 </tr>
 </thead>
 <tbody>
 <tr>
 <td align="center"><video src="https://github.com/harry0703/MoneyPrinterTurbo/assets/4928832/a84d33d5-27a2-4aba-8fd0-9fb2bd91c6a6"></video></td>
+<td align="center"><video src="https://github.com/harry0703/MoneyPrinterTurbo/assets/4928832/af2f3b0b-002e-49fe-b161-18ba91c055e8"></video></td>
 <td align="center"><video src="https://github.com/harry0703/MoneyPrinterTurbo/assets/4928832/112c9564-d52b-4472-99ad-970b75f66476"></video></td>
 </tr>
 </tbody>
@@ -106,16 +115,30 @@
 </table>
 
 ## 配置要求 📦
+
 - 建议最低 CPU 4核或以上，内存 8G 或以上，显卡非必须
 - Windows 10 或 MacOS 11.0 以上系统
 
+## 快速开始 🚀
+
+下载一键启动包，解压直接使用（路径不要有 **中文** 和 **空格**）
+
+### Windows
+
+- 百度网盘: https://pan.baidu.com/s/1MzBmcLTmVWohPEp9ohvvzA?pwd=pdcu 提取码: pdcu
+
+下载后，建议先**双击执行** `update.bat` 更新到**最新代码**，然后双击 `start.bat` 启动
+
+启动后，会自动打开浏览器（如果打开是空白，建议换成 **Chrome** 或者 **Edge** 打开）
+
+### 其他系统
+
+还没有制作一键启动包，看下面的 **安装部署** 部分，建议使用 **docker** 部署，更加方便。
+
 ## 安装部署 📥
 
-> 不想部署的可以直接下载安装包，解压直接使用
-- **Windows** 版本下载地址
-  - 百度网盘: https://pan.baidu.com/s/1BB3SGtAFTytzFLS5t2d8Gg?pwd=5bry
-
 ### 前提条件
+
 - 尽量不要使用 **中文路径**，避免出现一些无法预料的问题
 - 请确保你的 **网络** 是正常的，VPN需要打开`全局流量`模式
 
@@ -131,14 +154,6 @@ git clone https://github.com/harry0703/MoneyPrinterTurbo.git
 - 按照 `config.toml` 文件中的说明，配置好 `pexels_api_keys` 和 `llm_provider`，并根据 llm_provider 对应的服务商，配置相关的
   API Key
 
-#### ③ 配置大模型(LLM)
-
-- 如果要使用 `GPT-4.0` 或 `GPT-3.5`，需要有 `OpenAI` 的 `API Key`，如果没有，可以将 `llm_provider` 设置为 `g4f` (
-  一个免费使用GPT的开源库 https://github.com/xtekky/gpt4free ，但是该免费的服务，稳定性较差，有时候可以用，有时候用不了)
-- 或者可以使用到 [月之暗面](https://platform.moonshot.cn/console/api-keys) 申请。注册就送
-  15元体验金，可以对话1500次左右。然后设置 `llm_provider="moonshot"` 和 `moonshot_api_key`
-- 也可以使用 通义千问，具体请看配置文件里面的注释说明
-
 ### Docker部署 🐳
 
 #### ① 启动Docker
@@ -146,6 +161,7 @@ git clone https://github.com/harry0703/MoneyPrinterTurbo.git
 如果未安装 Docker，请先安装 https://www.docker.com/products/docker-desktop/
 
 如果是Windows系统，请参考微软的文档：
+
 1. https://learn.microsoft.com/zh-cn/windows/wsl/install
 2. https://learn.microsoft.com/zh-cn/windows/wsl/tutorials/wsl-containers
 
@@ -183,29 +199,24 @@ pip install -r requirements.txt
 
 #### ② 安装好 ImageMagick
 
-###### Windows:
+- Windows:
+    - 下载 https://imagemagick.org/script/download.php 选择Windows版本，切记一定要选择 **静态库** 版本，比如
+      ImageMagick-7.1.1-32-Q16-x64-**static**.exe
+    - 安装下载好的 ImageMagick，**注意不要修改安装路径**
+    - 修改 `配置文件 config.toml` 中的 `imagemagick_path` 为你的 **实际安装路径**
 
-- 下载 https://imagemagick.org/archive/binaries/ImageMagick-7.1.1-30-Q16-x64-static.exe
-- 安装下载好的 ImageMagick，注意不要修改安装路径
-- 修改 `配置文件 config.toml` 中的 `imagemagick_path` 为你的实际安装路径（如果安装的时候没有修改路径，直接取消注释即可）
-
-###### MacOS:
-
-```shell
-brew install imagemagick
-````
-
-###### Ubuntu
-
-```shell
-sudo apt-get install imagemagick
-```
-
-###### CentOS
-
-```shell
-sudo yum install ImageMagick
-```
+- MacOS:
+  ```shell
+  brew install imagemagick
+  ````
+- Ubuntu
+  ```shell
+  sudo apt-get install imagemagick
+  ```
+- CentOS
+  ```shell
+  sudo yum install ImageMagick
+  ```
 
 #### ③ 启动Web界面 🌐
 
@@ -224,7 +235,8 @@ webui.bat
 conda activate MoneyPrinterTurbo
 sh webui.sh
 ```
-启动后，会自动打开浏览器
+
+启动后，会自动打开浏览器（如果打开是空白，建议换成 **Chrome** 或者 **Edge** 打开）
 
 #### ④ 启动API服务 🚀
 
@@ -238,6 +250,8 @@ python main.py
 
 所有支持的声音列表，可以查看：[声音列表](./docs/voice-list.txt)
 
+2024-04-16 v1.1.2 新增了9种Azure的语音合成声音，需要配置API KEY，该声音合成的更加真实。
+
 ## 字幕生成 📜
 
 当前支持2种字幕生成方式：
@@ -250,17 +264,20 @@ python main.py
 建议使用 `edge` 模式，如果生成的字幕质量不好，再切换到 `whisper` 模式
 
 > 注意：
+
 1. whisper 模式下需要到 HuggingFace 下载一个模型文件，大约 3GB 左右，请确保网络通畅
 2. 如果留空，表示不生成字幕。
 
 > 由于国内无法访问 HuggingFace，可以使用以下方法下载 `whisper-large-v3` 的模型文件
 
 下载地址：
+
 - 百度网盘: https://pan.baidu.com/s/11h3Q6tsDtjQKTjUu3sc5cA?pwd=xjs9
 - 夸克网盘：https://pan.quark.cn/s/3ee3d991d64b
 
 模型下载后解压，整个目录放到 `.\MoneyPrinterTurbo\models` 里面，
 最终的文件路径应该是这样: `.\MoneyPrinterTurbo\models\whisper-large-v3`
+
 ```
 MoneyPrinterTurbo  
   ├─models
@@ -284,25 +301,32 @@ MoneyPrinterTurbo
 ## 常见问题 🤔
 
 ### ❓如何使用免费的OpenAI GPT-3.5模型?
+
 [OpenAI宣布ChatGPT里面3.5已经免费了](https://openai.com/blog/start-using-chatgpt-instantly)，有开发者将其封装成了API，可以直接调用
 
 **确保你安装和启动了docker服务**，执行以下命令启动docker服务
+
 ```shell
 docker run -p 3040:3040 missuo/freegpt35
 ```
+
 启动成功后，修改 `config.toml` 中的配置
+
 - `llm_provider` 设置为 `openai`
 - `openai_api_key` 随便填写一个即可，比如 '123456'
 - `openai_base_url` 改为 `http://localhost:3040/v1/`
 - `openai_model_name` 改为 `gpt-3.5-turbo`
 
+> 注意：该方式稳定性较差
 
 ### ❓AttributeError: 'str' object has no attribute 'choices'`
 
-这个问题是由于 OpenAI 或者其他 LLM，没有返回正确的回复导致的。
+这个问题是由于大模型没有返回正确的回复导致的。
 
 大概率是网络原因， 使用 **VPN**，或者设置 `openai_base_url` 为你的代理 ，应该就可以解决了。
 
+同时建议使用 **Moonshot** 或 **DeepSeek** 作为大模型提供商，这两个服务商在国内访问速度更快，更加稳定。
+
 ### ❓RuntimeError: No ffmpeg exe could be found
 
 通常情况下，ffmpeg 会被自动下载，并且会被自动检测到。
@@ -321,52 +345,14 @@ Install ffmpeg on your system, or set the IMAGEIO_FFMPEG_EXE environment variabl
 ffmpeg_path = "C:\\Users\\harry\\Downloads\\ffmpeg.exe"
 ```
 
-### ❓生成音频时报错或下载视频报错
-
-[issue 56](https://github.com/harry0703/MoneyPrinterTurbo/issues/56)
-
-```
-failed to generate audio, maybe the network is not available. 
-if you are in China, please use a VPN.
-```
-
-[issue 44](https://github.com/harry0703/MoneyPrinterTurbo/issues/44)
-
-```
-failed to download videos, maybe the network is not available. 
-if you are in China, please use a VPN.
-```
-
-这个大概率是网络原因，无法访问境外的服务，请使用VPN解决。
-
-### ❓ImageMagick is not installed on your computer
-
-[issue 33](https://github.com/harry0703/MoneyPrinterTurbo/issues/33)
-
-1. 按照 `示例配置` 里面提供的 `下载地址`
-   ，安装 https://imagemagick.org/archive/binaries/ImageMagick-7.1.1-29-Q16-x64-static.exe, 用静态库
-2. 不要安装在中文路径里面，避免出现一些无法预料的问题
-
-[issue 54](https://github.com/harry0703/MoneyPrinterTurbo/issues/54#issuecomment-2017842022)
-
-如果是linux系统，可以手动安装，参考 https://cn.linux-console.net/?p=16978
-
-感谢 [@wangwenqiao666](https://github.com/wangwenqiao666)的研究探索
-
 ### ❓ImageMagick的安全策略阻止了与临时文件@/tmp/tmpur5hyyto.txt相关的操作
 
-[issue 92](https://github.com/harry0703/MoneyPrinterTurbo/issues/92)
-
 可以在ImageMagick的配置文件policy.xml中找到这些策略。
 这个文件通常位于 /etc/ImageMagick-`X`/ 或 ImageMagick 安装目录的类似位置。
 修改包含`pattern="@"`的条目，将`rights="none"`更改为`rights="read|write"`以允许对文件的读写操作。
 
-感谢 [@chenhengzh](https://github.com/chenhengzh)的研究探索
-
 ### ❓OSError: [Errno 24] Too many open files
 
-[issue 100](https://github.com/harry0703/MoneyPrinterTurbo/issues/100)
-
 这个问题是由于系统打开文件数限制导致的，可以通过修改系统的文件打开数限制来解决。
 
 查看当前限制
@@ -381,26 +367,20 @@ ulimit -n
 ulimit -n 10240
 ```
 
-### ❓AttributeError: module 'PIL.Image' has no attribute 'ANTIALIAS'
+### ❓Whisper 模型下载失败，出现如下错误
 
-[issue 101](https://github.com/harry0703/MoneyPrinterTurbo/issues/101),
-[issue 83](https://github.com/harry0703/MoneyPrinterTurbo/issues/83),
-[issue 70](https://github.com/harry0703/MoneyPrinterTurbo/issues/70)
+LocalEntryNotfoundEror: Cannot find an appropriate cached snapshotfolderfor the specified revision on the local disk and
+outgoing trafic has been disabled.
+To enablerepo look-ups and downloads online, pass 'local files only=False' as input.
 
-先看下当前的 Pillow 版本是多少
+或者
 
-```shell
-pip list |grep Pillow
-```
+An error occured while synchronizing the model Systran/faster-whisper-large-v3 from the Hugging Face Hub:
+An error happened while trying to locate the files on the Hub and we cannot find the appropriate snapshot folder for the
+specified revision on the local disk. Please check your internet connection and try again.
+Trying to load the model directly from the local cache, if it exists.
 
-如果是 10.x 的版本，可以尝试下降级看看，有用户反馈降级后正常
-
-```shell
-pip uninstall Pillow
-pip install Pillow==9.5.0
-# 或者降级到 8.4.0
-pip install Pillow==8.4.0
-```
+解决方法：[点击查看如何从网盘手动下载模型](#%E5%AD%97%E5%B9%95%E7%94%9F%E6%88%90-)
 
 ## 反馈建议 📢
 
@@ -416,7 +396,6 @@ pip install Pillow==8.4.0
 
 点击查看 [`LICENSE`](LICENSE) 文件
 
-
 ## Star History
 
 [![Star History Chart](https://api.star-history.com/svg?repos=harry0703/MoneyPrinterTurbo&type=Date)](https://star-history.com/#harry0703/MoneyPrinterTurbo&Date)

app/asgi.py

@@ -1,4 +1,5 @@
 """Application implementation - ASGI."""
+
 import os
 
 from fastapi import FastAPI, Request
@@ -24,7 +25,9 @@ def exception_handler(request: Request, e: HttpException):
 def validation_exception_handler(request: Request, e: RequestValidationError):
     return JSONResponse(
         status_code=400,
-        content=utils.get_response(status=400, data=e.errors(), message='field required'),
+        content=utils.get_response(
+            status=400, data=e.errors(), message="field required"
+        ),
     )
 
 
@@ -61,7 +64,9 @@ app.add_middleware(
 )
 
 task_dir = utils.task_dir()
-app.mount("/tasks", StaticFiles(directory=task_dir, html=True, follow_symlink=True), name="")
+app.mount(
+    "/tasks", StaticFiles(directory=task_dir, html=True, follow_symlink=True), name=""
+)
 
 public_dir = utils.public_dir()
 app.mount("/", StaticFiles(directory=public_dir, html=True), name="")

app/config/__init__.py

@@ -10,7 +10,9 @@ from app.utils import utils
 def __init_logger():
     # _log_file = utils.storage_dir("logs/server.log")
     _lvl = config.log_level
-    root_dir = os.path.dirname(os.path.dirname(os.path.dirname(os.path.realpath(__file__))))
+    root_dir = os.path.dirname(
+        os.path.dirname(os.path.dirname(os.path.realpath(__file__)))
+    )
 
     def format_record(record):
         # 获取日志记录中的文件全路径
@@ -21,10 +23,13 @@ def __init_logger():
         record["file"].path = f"./{relative_path}"
         # 返回修改后的格式字符串
         # 您可以根据需要调整这里的格式
-        _format = '<green>{time:%Y-%m-%d %H:%M:%S}</> | ' + \
-                  '<level>{level}</> | ' + \
-                  '"{file.path}:{line}":<blue> {function}</> ' + \
-                  '- <level>{message}</>' + "\n"
+        _format = (
+            "<green>{time:%Y-%m-%d %H:%M:%S}</> | "
+            + "<level>{level}</> | "
+            + '"{file.path}:{line}":<blue> {function}</> '
+            + "- <level>{message}</>"
+            + "\n"
+        )
         return _format
 
     logger.remove()

app/config/config.py

@@ -25,7 +25,7 @@ def load_config():
         _config_ = toml.load(config_file)
     except Exception as e:
         logger.warning(f"load config failed: {str(e)}, try to load as utf-8-sig")
-        with open(config_file, mode="r", encoding='utf-8-sig') as fp:
+        with open(config_file, mode="r", encoding="utf-8-sig") as fp:
             _cfg_content = fp.read()
             _config_ = toml.loads(_cfg_content)
     return _config_
@@ -34,8 +34,6 @@ def load_config():
 def save_config():
     with open(config_file, "w", encoding="utf-8") as f:
         _cfg["app"] = app
-        _cfg["whisper"] = whisper
-        _cfg["pexels"] = pexels
         _cfg["azure"] = azure
         _cfg["ui"] = ui
         f.write(toml.dumps(_cfg))
@@ -44,7 +42,7 @@ def save_config():
 _cfg = load_config()
 app = _cfg.get("app", {})
 whisper = _cfg.get("whisper", {})
-pexels = _cfg.get("pexels", {})
+proxy = _cfg.get("proxy", {})
 azure = _cfg.get("azure", {})
 ui = _cfg.get("ui", {})
 
@@ -54,9 +52,11 @@ log_level = _cfg.get("log_level", "DEBUG")
 listen_host = _cfg.get("listen_host", "0.0.0.0")
 listen_port = _cfg.get("listen_port", 8080)
 project_name = _cfg.get("project_name", "MoneyPrinterTurbo")
-project_description = _cfg.get("project_description",
-                               "<a href='https://github.com/harry0703/MoneyPrinterTurbo'>https://github.com/harry0703/MoneyPrinterTurbo</a>")
-project_version = _cfg.get("project_version", "1.1.2")
+project_description = _cfg.get(
+    "project_description",
+    "<a href='https://github.com/harry0703/MoneyPrinterTurbo'>https://github.com/harry0703/MoneyPrinterTurbo</a>",
+)
+project_version = _cfg.get("project_version", "1.2.0")
 reload_debug = False
 
 imagemagick_path = app.get("imagemagick_path", "")

app/controllers/base.py

@@ -7,14 +7,14 @@ from app.models.exception import HttpException
 
 
 def get_task_id(request: Request):
-    task_id = request.headers.get('x-task-id')
+    task_id = request.headers.get("x-task-id")
     if not task_id:
         task_id = uuid4()
     return str(task_id)
 
 
 def get_api_key(request: Request):
-    api_key = request.headers.get('x-api-key')
+    api_key = request.headers.get("x-api-key")
     return api_key
 
 
@@ -23,5 +23,9 @@ def verify_token(request: Request):
     if token != config.app.get("api_key", ""):
         request_id = get_task_id(request)
         request_url = request.url
-        user_agent = request.headers.get('user-agent')
-        raise HttpException(task_id=request_id, status_code=401, message=f"invalid token: {request_url}, {user_agent}")
+        user_agent = request.headers.get("user-agent")
+        raise HttpException(
+            task_id=request_id,
+            status_code=401,
+            message=f"invalid token: {request_url}, {user_agent}",
+        )

app/controllers/manager/base_manager.py

@@ -0,0 +1,64 @@
+import threading
+from typing import Callable, Any, Dict
+
+
+class TaskManager:
+    def __init__(self, max_concurrent_tasks: int):
+        self.max_concurrent_tasks = max_concurrent_tasks
+        self.current_tasks = 0
+        self.lock = threading.Lock()
+        self.queue = self.create_queue()
+
+    def create_queue(self):
+        raise NotImplementedError()
+
+    def add_task(self, func: Callable, *args: Any, **kwargs: Any):
+        with self.lock:
+            if self.current_tasks < self.max_concurrent_tasks:
+                print(f"add task: {func.__name__}, current_tasks: {self.current_tasks}")
+                self.execute_task(func, *args, **kwargs)
+            else:
+                print(
+                    f"enqueue task: {func.__name__}, current_tasks: {self.current_tasks}"
+                )
+                self.enqueue({"func": func, "args": args, "kwargs": kwargs})
+
+    def execute_task(self, func: Callable, *args: Any, **kwargs: Any):
+        thread = threading.Thread(
+            target=self.run_task, args=(func, *args), kwargs=kwargs
+        )
+        thread.start()
+
+    def run_task(self, func: Callable, *args: Any, **kwargs: Any):
+        try:
+            with self.lock:
+                self.current_tasks += 1
+            func(*args, **kwargs)  # 在这里调用函数，传递*args和**kwargs
+        finally:
+            self.task_done()
+
+    def check_queue(self):
+        with self.lock:
+            if (
+                self.current_tasks < self.max_concurrent_tasks
+                and not self.is_queue_empty()
+            ):
+                task_info = self.dequeue()
+                func = task_info["func"]
+                args = task_info.get("args", ())
+                kwargs = task_info.get("kwargs", {})
+                self.execute_task(func, *args, **kwargs)
+
+    def task_done(self):
+        with self.lock:
+            self.current_tasks -= 1
+        self.check_queue()
+
+    def enqueue(self, task: Dict):
+        raise NotImplementedError()
+
+    def dequeue(self):
+        raise NotImplementedError()
+
+    def is_queue_empty(self):
+        raise NotImplementedError()

app/controllers/manager/memory_manager.py

@@ -0,0 +1,18 @@
+from queue import Queue
+from typing import Dict
+
+from app.controllers.manager.base_manager import TaskManager
+
+
+class InMemoryTaskManager(TaskManager):
+    def create_queue(self):
+        return Queue()
+
+    def enqueue(self, task: Dict):
+        self.queue.put(task)
+
+    def dequeue(self):
+        return self.queue.get()
+
+    def is_queue_empty(self):
+        return self.queue.empty()

app/controllers/manager/redis_manager.py

@@ -0,0 +1,56 @@
+import json
+from typing import Dict
+
+import redis
+
+from app.controllers.manager.base_manager import TaskManager
+from app.models.schema import VideoParams
+from app.services import task as tm
+
+FUNC_MAP = {
+    "start": tm.start,
+    # 'start_test': tm.start_test
+}
+
+
+class RedisTaskManager(TaskManager):
+    def __init__(self, max_concurrent_tasks: int, redis_url: str):
+        self.redis_client = redis.Redis.from_url(redis_url)
+        super().__init__(max_concurrent_tasks)
+
+    def create_queue(self):
+        return "task_queue"
+
+    def enqueue(self, task: Dict):
+        task_with_serializable_params = task.copy()
+
+        if "params" in task["kwargs"] and isinstance(
+            task["kwargs"]["params"], VideoParams
+        ):
+            task_with_serializable_params["kwargs"]["params"] = task["kwargs"][
+                "params"
+            ].dict()
+
+        # 将函数对象转换为其名称
+        task_with_serializable_params["func"] = task["func"].__name__
+        self.redis_client.rpush(self.queue, json.dumps(task_with_serializable_params))
+
+    def dequeue(self):
+        task_json = self.redis_client.lpop(self.queue)
+        if task_json:
+            task_info = json.loads(task_json)
+            # 将函数名称转换回函数对象
+            task_info["func"] = FUNC_MAP[task_info["func"]]
+
+            if "params" in task_info["kwargs"] and isinstance(
+                task_info["kwargs"]["params"], dict
+            ):
+                task_info["kwargs"]["params"] = VideoParams(
+                    **task_info["kwargs"]["params"]
+                )
+
+            return task_info
+        return None
+
+    def is_queue_empty(self):
+        return self.redis_client.llen(self.queue) == 0

app/controllers/ping.py

@@ -4,6 +4,11 @@ from fastapi import Request
 router = APIRouter()
 
 
-@router.get("/ping", tags=["Health Check"], description="检查服务可用性", response_description="pong")
+@router.get(
+    "/ping",
+    tags=["Health Check"],
+    description="检查服务可用性",
+    response_description="pong",
+)
 def ping(request: Request) -> str:
     return "pong"

app/controllers/v1/base.py

@@ -3,8 +3,8 @@ from fastapi import APIRouter, Depends
 
 def new_router(dependencies=None):
     router = APIRouter()
-    router.tags = ['V1']
-    router.prefix = '/api/v1'
+    router.tags = ["V1"]
+    router.prefix = "/api/v1"
     # 将认证依赖项应用于所有路由
     if dependencies:
         router.dependencies = dependencies

app/controllers/v1/llm.py

@@ -1,6 +1,11 @@
 from fastapi import Request
 from app.controllers.v1.base import new_router
-from app.models.schema import VideoScriptResponse, VideoScriptRequest, VideoTermsResponse, VideoTermsRequest
+from app.models.schema import (
+    VideoScriptResponse,
+    VideoScriptRequest,
+    VideoTermsResponse,
+    VideoTermsRequest,
+)
 from app.services import llm
 from app.utils import utils
 
@@ -9,23 +14,31 @@ from app.utils import utils
 router = new_router()
 
 
-@router.post("/scripts", response_model=VideoScriptResponse, summary="Create a script for the video")
+@router.post(
+    "/scripts",
+    response_model=VideoScriptResponse,
+    summary="Create a script for the video",
+)
 def generate_video_script(request: Request, body: VideoScriptRequest):
-    video_script = llm.generate_script(video_subject=body.video_subject,
-                                       language=body.video_language,
-                                       paragraph_number=body.paragraph_number)
-    response = {
-        "video_script": video_script
-    }
+    video_script = llm.generate_script(
+        video_subject=body.video_subject,
+        language=body.video_language,
+        paragraph_number=body.paragraph_number,
+    )
+    response = {"video_script": video_script}
     return utils.get_response(200, response)
 
 
-@router.post("/terms", response_model=VideoTermsResponse, summary="Generate video terms based on the video script")
+@router.post(
+    "/terms",
+    response_model=VideoTermsResponse,
+    summary="Generate video terms based on the video script",
+)
 def generate_video_terms(request: Request, body: VideoTermsRequest):
-    video_terms = llm.generate_terms(video_subject=body.video_subject,
-                                     video_script=body.video_script,
-                                     amount=body.amount)
-    response = {
-        "video_terms": video_terms
-    }
+    video_terms = llm.generate_terms(
+        video_subject=body.video_subject,
+        video_script=body.video_script,
+        amount=body.amount,
+    )
+    response = {"video_terms": video_terms}
     return utils.get_response(200, response)

app/controllers/v1/video.py

@@ -1,49 +1,108 @@
-import os
 import glob
+import os
 import pathlib
 import shutil
+from typing import Union
 
-from fastapi import Request, Depends, Path, BackgroundTasks, UploadFile
-from fastapi.responses import FileResponse, StreamingResponse
+from fastapi import BackgroundTasks, Depends, Path, Request, UploadFile
 from fastapi.params import File
+from fastapi.responses import FileResponse, StreamingResponse
 from loguru import logger
 
 from app.config import config
 from app.controllers import base
+from app.controllers.manager.memory_manager import InMemoryTaskManager
+from app.controllers.manager.redis_manager import RedisTaskManager
 from app.controllers.v1.base import new_router
 from app.models.exception import HttpException
-from app.models.schema import TaskVideoRequest, TaskQueryResponse, TaskResponse, TaskQueryRequest, \
-    BgmUploadResponse, BgmRetrieveResponse, TaskDeletionResponse
-from app.services import task as tm
+from app.models.schema import (
+    AudioRequest,
+    BgmRetrieveResponse,
+    BgmUploadResponse,
+    SubtitleRequest,
+    TaskDeletionResponse,
+    TaskQueryRequest,
+    TaskQueryResponse,
+    TaskResponse,
+    TaskVideoRequest,
+)
 from app.services import state as sm
+from app.services import task as tm
 from app.utils import utils
 
 # 认证依赖项
 # router = new_router(dependencies=[Depends(base.verify_token)])
 router = new_router()
 
+_enable_redis = config.app.get("enable_redis", False)
+_redis_host = config.app.get("redis_host", "localhost")
+_redis_port = config.app.get("redis_port", 6379)
+_redis_db = config.app.get("redis_db", 0)
+_redis_password = config.app.get("redis_password", None)
+_max_concurrent_tasks = config.app.get("max_concurrent_tasks", 5)
+
+redis_url = f"redis://:{_redis_password}@{_redis_host}:{_redis_port}/{_redis_db}"
+# 根据配置选择合适的任务管理器
+if _enable_redis:
+    task_manager = RedisTaskManager(
+        max_concurrent_tasks=_max_concurrent_tasks, redis_url=redis_url
+    )
+else:
+    task_manager = InMemoryTaskManager(max_concurrent_tasks=_max_concurrent_tasks)
+
 
 @router.post("/videos", response_model=TaskResponse, summary="Generate a short video")
-def create_video(background_tasks: BackgroundTasks, request: Request, body: TaskVideoRequest):
+def create_video(
+    background_tasks: BackgroundTasks, request: Request, body: TaskVideoRequest
+):
+    return create_task(request, body, stop_at="video")
+
+
+@router.post("/subtitle", response_model=TaskResponse, summary="Generate subtitle only")
+def create_subtitle(
+    background_tasks: BackgroundTasks, request: Request, body: SubtitleRequest
+):
+    return create_task(request, body, stop_at="subtitle")
+
+
+@router.post("/audio", response_model=TaskResponse, summary="Generate audio only")
+def create_audio(
+    background_tasks: BackgroundTasks, request: Request, body: AudioRequest
+):
+    return create_task(request, body, stop_at="audio")
+
+
+def create_task(
+    request: Request,
+    body: Union[TaskVideoRequest, SubtitleRequest, AudioRequest],
+    stop_at: str,
+):
     task_id = utils.get_uuid()
     request_id = base.get_task_id(request)
     try:
         task = {
             "task_id": task_id,
             "request_id": request_id,
-            "params": body.dict(),
+            "params": body.model_dump(),
         }
         sm.state.update_task(task_id)
-        background_tasks.add_task(tm.start, task_id=task_id, params=body)
-        logger.success(f"video created: {utils.to_json(task)}")
+        task_manager.add_task(tm.start, task_id=task_id, params=body, stop_at=stop_at)
+        logger.success(f"Task created: {utils.to_json(task)}")
         return utils.get_response(200, task)
     except ValueError as e:
-        raise HttpException(task_id=task_id, status_code=400, message=f"{request_id}: {str(e)}")
+        raise HttpException(
+            task_id=task_id, status_code=400, message=f"{request_id}: {str(e)}"
+        )
 
 
-@router.get("/tasks/{task_id}", response_model=TaskQueryResponse, summary="Query task status")
-def get_task(request: Request, task_id: str = Path(..., description="Task ID"),
-             query: TaskQueryRequest = Depends()):
+@router.get(
+    "/tasks/{task_id}", response_model=TaskQueryResponse, summary="Query task status"
+)
+def get_task(
+    request: Request,
+    task_id: str = Path(..., description="Task ID"),
+    query: TaskQueryRequest = Depends(),
+):
     endpoint = config.app.get("endpoint", "")
     if not endpoint:
         endpoint = str(request.base_url)
@@ -76,10 +135,16 @@ def get_task(request: Request, task_id: str = Path(..., description="Task ID"),
             task["combined_videos"] = urls
         return utils.get_response(200, task)
 
-    raise HttpException(task_id=task_id, status_code=404, message=f"{request_id}: task not found")
+    raise HttpException(
+        task_id=task_id, status_code=404, message=f"{request_id}: task not found"
+    )
 
 
-@router.delete("/tasks/{task_id}", response_model=TaskDeletionResponse, summary="Delete a generated short video task")
+@router.delete(
+    "/tasks/{task_id}",
+    response_model=TaskDeletionResponse,
+    summary="Delete a generated short video task",
+)
 def delete_video(request: Request, task_id: str = Path(..., description="Task ID")):
     request_id = base.get_task_id(request)
     task = sm.state.get_task(task_id)
@@ -93,32 +158,40 @@ def delete_video(request: Request, task_id: str = Path(..., description="Task ID
         logger.success(f"video deleted: {utils.to_json(task)}")
         return utils.get_response(200)
 
-    raise HttpException(task_id=task_id, status_code=404, message=f"{request_id}: task not found")
+    raise HttpException(
+        task_id=task_id, status_code=404, message=f"{request_id}: task not found"
+    )
 
 
-@router.get("/musics", response_model=BgmRetrieveResponse, summary="Retrieve local BGM files")
+@router.get(
+    "/musics", response_model=BgmRetrieveResponse, summary="Retrieve local BGM files"
+)
 def get_bgm_list(request: Request):
     suffix = "*.mp3"
     song_dir = utils.song_dir()
     files = glob.glob(os.path.join(song_dir, suffix))
     bgm_list = []
     for file in files:
-        bgm_list.append({
-            "name": os.path.basename(file),
-            "size": os.path.getsize(file),
-            "file": file,
-        })
-    response = {
-        "files": bgm_list
-    }
+        bgm_list.append(
+            {
+                "name": os.path.basename(file),
+                "size": os.path.getsize(file),
+                "file": file,
+            }
+        )
+    response = {"files": bgm_list}
     return utils.get_response(200, response)
 
 
-@router.post("/musics", response_model=BgmUploadResponse, summary="Upload the BGM file to the songs directory")
+@router.post(
+    "/musics",
+    response_model=BgmUploadResponse,
+    summary="Upload the BGM file to the songs directory",
+)
 def upload_bgm_file(request: Request, file: UploadFile = File(...)):
     request_id = base.get_task_id(request)
     # check file ext
-    if file.filename.endswith('mp3'):
+    if file.filename.endswith("mp3"):
         song_dir = utils.song_dir()
         save_path = os.path.join(song_dir, file.filename)
         # save file
@@ -126,26 +199,26 @@ def upload_bgm_file(request: Request, file: UploadFile = File(...)):
             # If the file already exists, it will be overwritten
             file.file.seek(0)
             buffer.write(file.file.read())
-        response = {
-            "file": save_path
-        }
+        response = {"file": save_path}
         return utils.get_response(200, response)
 
-    raise HttpException('', status_code=400, message=f"{request_id}: Only *.mp3 files can be uploaded")
+    raise HttpException(
+        "", status_code=400, message=f"{request_id}: Only *.mp3 files can be uploaded"
+    )
 
 
 @router.get("/stream/{file_path:path}")
 async def stream_video(request: Request, file_path: str):
     tasks_dir = utils.task_dir()
     video_path = os.path.join(tasks_dir, file_path)
-    range_header = request.headers.get('Range')
+    range_header = request.headers.get("Range")
     video_size = os.path.getsize(video_path)
     start, end = 0, video_size - 1
 
     length = video_size
     if range_header:
-        range_ = range_header.split('bytes=')[1]
-        start, end = [int(part) if part else None for part in range_.split('-')]
+        range_ = range_header.split("bytes=")[1]
+        start, end = [int(part) if part else None for part in range_.split("-")]
         if start is None:
             start = video_size - end
             end = video_size - 1
@@ -154,7 +227,7 @@ async def stream_video(request: Request, file_path: str):
         length = end - start + 1
 
     def file_iterator(file_path, offset=0, bytes_to_read=None):
-        with open(file_path, 'rb') as f:
+        with open(file_path, "rb") as f:
             f.seek(offset, os.SEEK_SET)
             remaining = bytes_to_read or video_size
             while remaining > 0:
@@ -165,10 +238,12 @@ async def stream_video(request: Request, file_path: str):
                 remaining -= len(data)
                 yield data
 
-    response = StreamingResponse(file_iterator(video_path, start, length), media_type='video/mp4')
-    response.headers['Content-Range'] = f'bytes {start}-{end}/{video_size}'
-    response.headers['Accept-Ranges'] = 'bytes'
-    response.headers['Content-Length'] = str(length)
+    response = StreamingResponse(
+        file_iterator(video_path, start, length), media_type="video/mp4"
+    )
+    response.headers["Content-Range"] = f"bytes {start}-{end}/{video_size}"
+    response.headers["Accept-Ranges"] = "bytes"
+    response.headers["Content-Length"] = str(length)
     response.status_code = 206  # Partial Content
 
     return response
@@ -187,8 +262,10 @@ async def download_video(_: Request, file_path: str):
     file_path = pathlib.Path(video_path)
     filename = file_path.stem
     extension = file_path.suffix
-    headers = {
-        "Content-Disposition": f"attachment; filename={filename}{extension}"
-    }
-    return FileResponse(path=video_path, headers=headers, filename=f"{filename}{extension}",
-                        media_type=f'video/{extension[1:]}')
+    headers = {"Content-Disposition": f"attachment; filename={filename}{extension}"}
+    return FileResponse(
+        path=video_path,
+        headers=headers,
+        filename=f"{filename}{extension}",
+        media_type=f"video/{extension[1:]}",
+    )

app/models/const.py

@@ -1,8 +1,25 @@
 PUNCTUATIONS = [
-    "?", ",", ".", "、", ";", ":", "!", "…",
-    "？", "，", "。", "、", "；", "：", "！", "...",
+    "?",
+    ",",
+    ".",
+    "、",
+    ";",
+    ":",
+    "!",
+    "…",
+    "？",
+    "，",
+    "。",
+    "、",
+    "；",
+    "：",
+    "！",
+    "...",
 ]
 
 TASK_STATE_FAILED = -1
 TASK_STATE_COMPLETE = 1
 TASK_STATE_PROCESSING = 4
+
+FILE_TYPE_VIDEOS = ["mp4", "mov", "mkv", "webm"]
+FILE_TYPE_IMAGES = ["jpg", "jpeg", "png", "bmp"]

app/models/exception.py

@@ -5,16 +5,18 @@ from loguru import logger
 
 
 class HttpException(Exception):
-    def __init__(self, task_id: str, status_code: int, message: str = '', data: Any = None):
+    def __init__(
+        self, task_id: str, status_code: int, message: str = "", data: Any = None
+    ):
         self.message = message
         self.status_code = status_code
         self.data = data
         # 获取异常堆栈信息
         tb_str = traceback.format_exc().strip()
         if not tb_str or tb_str == "NoneType: None":
-            msg = f'HttpException: {status_code}, {task_id}, {message}'
+            msg = f"HttpException: {status_code}, {task_id}, {message}"
         else:
-            msg = f'HttpException: {status_code}, {task_id}, {message}\n{tb_str}'
+            msg = f"HttpException: {status_code}, {task_id}, {message}\n{tb_str}"
 
         if status_code == 400:
             logger.warning(msg)

app/models/schema.py

@@ -1,11 +1,16 @@
-from enum import Enum
-from typing import Any, Optional
-
-from pydantic import BaseModel
 import warnings
+from enum import Enum
+from typing import Any, List, Optional
+
+import pydantic
+from pydantic import BaseModel
 
 # 忽略 Pydantic 的特定警告
-warnings.filterwarnings("ignore", category=UserWarning, message="Field name.*shadows an attribute in parent.*")
+warnings.filterwarnings(
+    "ignore",
+    category=UserWarning,
+    message="Field name.*shadows an attribute in parent.*",
+)
 
 
 class VideoConcatMode(str, Enum):
@@ -28,6 +33,11 @@ class VideoAspect(str, Enum):
         return 1080, 1920
 
 
+class _Config:
+    arbitrary_types_allowed = True
+
+
+@pydantic.dataclasses.dataclass(config=_Config)
 class MaterialInfo:
     provider: str = "pexels"
     url: str = ""
@@ -55,7 +65,6 @@ class MaterialInfo:
 #     # "male-zh-TW-YunJheNeural",
 #
 #     # en-US
-#
 #     "female-en-US-AnaNeural",
 #     "female-en-US-AriaNeural",
 #     "female-en-US-AvaNeural",
@@ -73,7 +82,7 @@ class MaterialInfo:
 # ]
 
 
-class VideoParams:
+class VideoParams(BaseModel):
     """
     {
       "video_subject": "",
@@ -87,6 +96,7 @@ class VideoParams:
       "stroke_width": 1.5
     }
     """
+
     video_subject: str
     video_script: str = ""  # 用于生成视频的脚本
     video_terms: Optional[str | list] = None  # 用于生成视频的关键词
@@ -95,16 +105,21 @@ class VideoParams:
     video_clip_duration: Optional[int] = 5
     video_count: Optional[int] = 1
 
+    video_source: Optional[str] = "pexels"
+    video_materials: Optional[List[MaterialInfo]] = None  # 用于生成视频的素材
+
     video_language: Optional[str] = ""  # auto detect
 
     voice_name: Optional[str] = ""
     voice_volume: Optional[float] = 1.0
+    voice_rate: Optional[float] = 1.0
     bgm_type: Optional[str] = "random"
     bgm_file: Optional[str] = ""
     bgm_volume: Optional[float] = 0.2
 
     subtitle_enabled: Optional[bool] = True
     subtitle_position: Optional[str] = "bottom"  # top, bottom, center
+    custom_position: float = 70.0
     font_name: Optional[str] = "STHeitiMedium.ttc"
     text_fore_color: Optional[str] = "#FFFFFF"
     text_background_color: Optional[str] = "transparent"
@@ -116,6 +131,38 @@ class VideoParams:
     paragraph_number: Optional[int] = 1
 
 
+class SubtitleRequest(BaseModel):
+    video_script: str
+    video_language: Optional[str] = ""
+    voice_name: Optional[str] = "zh-CN-XiaoxiaoNeural-Female"
+    voice_volume: Optional[float] = 1.0
+    voice_rate: Optional[float] = 1.2
+    bgm_type: Optional[str] = "random"
+    bgm_file: Optional[str] = ""
+    bgm_volume: Optional[float] = 0.2
+    subtitle_position: Optional[str] = "bottom"
+    font_name: Optional[str] = "STHeitiMedium.ttc"
+    text_fore_color: Optional[str] = "#FFFFFF"
+    text_background_color: Optional[str] = "transparent"
+    font_size: int = 60
+    stroke_color: Optional[str] = "#000000"
+    stroke_width: float = 1.5
+    video_source: Optional[str] = "local"
+    subtitle_enabled: Optional[str] = "true"
+
+
+class AudioRequest(BaseModel):
+    video_script: str
+    video_language: Optional[str] = ""
+    voice_name: Optional[str] = "zh-CN-XiaoxiaoNeural-Female"
+    voice_volume: Optional[float] = 1.0
+    voice_rate: Optional[float] = 1.2
+    bgm_type: Optional[str] = "random"
+    bgm_file: Optional[str] = ""
+    bgm_volume: Optional[float] = 0.2
+    video_source: Optional[str] = "local"
+
+
 class VideoScriptParams:
     """
     {
@@ -124,6 +171,7 @@ class VideoScriptParams:
       "paragraph_number": 1
     }
     """
+
     video_subject: Optional[str] = "春天的花海"
     video_language: Optional[str] = ""
     paragraph_number: Optional[int] = 1
@@ -137,14 +185,17 @@ class VideoTermsParams:
       "amount": 5
     }
     """
+
     video_subject: Optional[str] = "春天的花海"
-    video_script: Optional[str] = "春天的花海，如诗如画般展现在眼前。万物复苏的季节里，大地披上了一袭绚丽多彩的盛装。金黄的迎春、粉嫩的樱花、洁白的梨花、艳丽的郁金香……"
+    video_script: Optional[str] = (
+        "春天的花海，如诗如画般展现在眼前。万物复苏的季节里，大地披上了一袭绚丽多彩的盛装。金黄的迎春、粉嫩的樱花、洁白的梨花、艳丽的郁金香……"
+    )
     amount: Optional[int] = 5
 
 
 class BaseResponse(BaseModel):
     status: int = 200
-    message: Optional[str] = 'success'
+    message: Optional[str] = "success"
     data: Any = None
 
 
@@ -179,9 +230,7 @@ class TaskResponse(BaseResponse):
             "example": {
                 "status": 200,
                 "message": "success",
-                "data": {
-                    "task_id": "6c85c8cc-a77a-42b9-bc30-947815aa0558"
-                }
+                "data": {"task_id": "6c85c8cc-a77a-42b9-bc30-947815aa0558"},
             },
         }
 
@@ -200,8 +249,8 @@ class TaskQueryResponse(BaseResponse):
                     ],
                     "combined_videos": [
                         "http://127.0.0.1:8080/tasks/6c85c8cc-a77a-42b9-bc30-947815aa0558/combined-1.mp4"
-                    ]
-                }
+                    ],
+                },
             },
         }
 
@@ -220,8 +269,8 @@ class TaskDeletionResponse(BaseResponse):
                     ],
                     "combined_videos": [
                         "http://127.0.0.1:8080/tasks/6c85c8cc-a77a-42b9-bc30-947815aa0558/combined-1.mp4"
-                    ]
-                }
+                    ],
+                },
             },
         }
 
@@ -234,7 +283,7 @@ class VideoScriptResponse(BaseResponse):
                 "message": "success",
                 "data": {
                     "video_script": "春天的花海，是大自然的一幅美丽画卷。在这个季节里，大地复苏，万物生长，花朵争相绽放，形成了一片五彩斑斓的花海..."
-                }
+                },
             },
         }
 
@@ -245,9 +294,7 @@ class VideoTermsResponse(BaseResponse):
             "example": {
                 "status": 200,
                 "message": "success",
-                "data": {
-                    "video_terms": ["sky", "tree"]
-                }
+                "data": {"video_terms": ["sky", "tree"]},
             },
         }
 
@@ -263,10 +310,10 @@ class BgmRetrieveResponse(BaseResponse):
                         {
                             "name": "output013.mp3",
                             "size": 1891269,
-                            "file": "/MoneyPrinterTurbo/resource/songs/output013.mp3"
+                            "file": "/MoneyPrinterTurbo/resource/songs/output013.mp3",
                         }
                     ]
-                }
+                },
             },
         }
 
@@ -277,8 +324,6 @@ class BgmUploadResponse(BaseResponse):
             "example": {
                 "status": 200,
                 "message": "success",
-                "data": {
-                    "file": "/MoneyPrinterTurbo/resource/songs/example.mp3"
-                }
+                "data": {"file": "/MoneyPrinterTurbo/resource/songs/example.mp3"},
             },
         }

app/router.py

@@ -6,9 +6,10 @@ Resources:
     1. https://fastapi.tiangolo.com/tutorial/bigger-applications
 
 """
+
 from fastapi import APIRouter
 
-from app.controllers.v1 import video, llm
+from app.controllers.v1 import llm, video
 
 root_api_router = APIRouter()
 # v1

app/services/llm.py

@@ -9,6 +9,8 @@ from openai.types.chat import ChatCompletion
 
 from app.config import config
 
+_max_retries = 5
+
 
 def _generate_response(prompt: str) -> str:
     content = ""
@@ -19,6 +21,7 @@ def _generate_response(prompt: str) -> str:
         if not model_name:
             model_name = "gpt-3.5-turbo-16k-0613"
         import g4f
+
         content = g4f.ChatCompletion.create(
             model=model_name,
             messages=[{"role": "user", "content": prompt}],
@@ -64,43 +67,68 @@ def _generate_response(prompt: str) -> str:
             model_name = config.app.get("cloudflare_model_name")
             account_id = config.app.get("cloudflare_account_id")
             base_url = "***"
+        elif llm_provider == "deepseek":
+            api_key = config.app.get("deepseek_api_key")
+            model_name = config.app.get("deepseek_model_name")
+            base_url = config.app.get("deepseek_base_url")
+            if not base_url:
+                base_url = "https://api.deepseek.com"
+        elif llm_provider == "ernie":
+            api_key = config.app.get("ernie_api_key")
+            secret_key = config.app.get("ernie_secret_key")
+            base_url = config.app.get("ernie_base_url")
+            model_name = "***"
+            if not secret_key:
+                raise ValueError(
+                    f"{llm_provider}: secret_key is not set, please set it in the config.toml file."
+                )
         else:
-            raise ValueError("llm_provider is not set, please set it in the config.toml file.")
+            raise ValueError(
+                "llm_provider is not set, please set it in the config.toml file."
+            )
 
         if not api_key:
-            raise ValueError(f"{llm_provider}: api_key is not set, please set it in the config.toml file.")
+            raise ValueError(
+                f"{llm_provider}: api_key is not set, please set it in the config.toml file."
+            )
         if not model_name:
-            raise ValueError(f"{llm_provider}: model_name is not set, please set it in the config.toml file.")
+            raise ValueError(
+                f"{llm_provider}: model_name is not set, please set it in the config.toml file."
+            )
         if not base_url:
-            raise ValueError(f"{llm_provider}: base_url is not set, please set it in the config.toml file.")
+            raise ValueError(
+                f"{llm_provider}: base_url is not set, please set it in the config.toml file."
+            )
 
         if llm_provider == "qwen":
             import dashscope
             from dashscope.api_entities.dashscope_response import GenerationResponse
+
             dashscope.api_key = api_key
             response = dashscope.Generation.call(
-                model=model_name,
-                messages=[{"role": "user", "content": prompt}]
+                model=model_name, messages=[{"role": "user", "content": prompt}]
             )
             if response:
                 if isinstance(response, GenerationResponse):
                     status_code = response.status_code
                     if status_code != 200:
                         raise Exception(
-                            f"[{llm_provider}] returned an error response: \"{response}\"")
+                            f'[{llm_provider}] returned an error response: "{response}"'
+                        )
 
                     content = response["output"]["text"]
                     return content.replace("\n", "")
                 else:
                     raise Exception(
-                        f"[{llm_provider}] returned an invalid response: \"{response}\"")
+                        f'[{llm_provider}] returned an invalid response: "{response}"'
+                    )
             else:
-                raise Exception(
-                    f"[{llm_provider}] returned an empty response")
+                raise Exception(f"[{llm_provider}] returned an empty response")
 
         if llm_provider == "gemini":
             import google.generativeai as genai
-            genai.configure(api_key=api_key, transport='rest')
+
+            genai.configure(api_key=api_key, transport="rest")
 
             generation_config = {
                 "temperature": 0.5,
@@ -112,25 +140,27 @@ def _generate_response(prompt: str) -> str:
             safety_settings = [
                 {
                     "category": "HARM_CATEGORY_HARASSMENT",
-                    "threshold": "BLOCK_ONLY_HIGH"
+                    "threshold": "BLOCK_ONLY_HIGH",
                 },
                 {
                     "category": "HARM_CATEGORY_HATE_SPEECH",
-                    "threshold": "BLOCK_ONLY_HIGH"
+                    "threshold": "BLOCK_ONLY_HIGH",
                 },
                 {
                     "category": "HARM_CATEGORY_SEXUALLY_EXPLICIT",
-                    "threshold": "BLOCK_ONLY_HIGH"
+                    "threshold": "BLOCK_ONLY_HIGH",
                 },
                 {
                     "category": "HARM_CATEGORY_DANGEROUS_CONTENT",
-                    "threshold": "BLOCK_ONLY_HIGH"
+                    "threshold": "BLOCK_ONLY_HIGH",
                 },
             ]
 
-            model = genai.GenerativeModel(model_name=model_name,
-                                          generation_config=generation_config,
-                                          safety_settings=safety_settings)
+            model = genai.GenerativeModel(
+                model_name=model_name,
+                generation_config=generation_config,
+                safety_settings=safety_settings,
+            )
 
             try:
                 response = model.generate_content(prompt)
@@ -143,20 +173,54 @@ def _generate_response(prompt: str) -> str:
 
         if llm_provider == "cloudflare":
             import requests
+
             response = requests.post(
                 f"https://api.cloudflare.com/client/v4/accounts/{account_id}/ai/run/{model_name}",
                 headers={"Authorization": f"Bearer {api_key}"},
                 json={
                     "messages": [
                         {"role": "system", "content": "You are a friendly assistant"},
-                        {"role": "user", "content": prompt}
+                        {"role": "user", "content": prompt},
                     ]
-                }
+                },
             )
             result = response.json()
             logger.info(result)
             return result["result"]["response"]
 
+        if llm_provider == "ernie":
+            import requests
+
+            params = {
+                "grant_type": "client_credentials",
+                "client_id": api_key,
+                "client_secret": secret_key,
+            }
+            access_token = (
+                requests.post("https://aip.baidubce.com/oauth/2.0/token", params=params)
+                .json()
+                .get("access_token")
+            )
+            url = f"{base_url}?access_token={access_token}"
+
+            payload = json.dumps(
+                {
+                    "messages": [{"role": "user", "content": prompt}],
+                    "temperature": 0.5,
+                    "top_p": 0.8,
+                    "penalty_score": 1,
+                    "disable_search": False,
+                    "enable_citation": False,
+                    "response_format": "text",
+                }
+            )
+            headers = {"Content-Type": "application/json"}
+
+            response = requests.request(
+                "POST", url, headers=headers, data=payload
+            ).json()
+            return response.get("result")
+
         if llm_provider == "azure":
             client = AzureOpenAI(
                 api_key=api_key,
@@ -170,24 +234,27 @@ def _generate_response(prompt: str) -> str:
             )
 
         response = client.chat.completions.create(
-            model=model_name,
-            messages=[{"role": "user", "content": prompt}]
+            model=model_name, messages=[{"role": "user", "content": prompt}]
         )
         if response:
             if isinstance(response, ChatCompletion):
                 content = response.choices[0].message.content
             else:
                 raise Exception(
-                    f"[{llm_provider}] returned an invalid response: \"{response}\", please check your network "
-                    f"connection and try again.")
+                    f'[{llm_provider}] returned an invalid response: "{response}", please check your network '
+                    f"connection and try again."
+                )
         else:
             raise Exception(
-                f"[{llm_provider}] returned an empty response, please check your network connection and try again.")
+                f"[{llm_provider}] returned an empty response, please check your network connection and try again."
+            )
 
     return content.replace("\n", "")
 
 
-def generate_script(video_subject: str, language: str = "", paragraph_number: int = 1) -> str:
+def generate_script(
+    video_subject: str, language: str = "", paragraph_number: int = 1
+) -> str:
     prompt = f"""
 # Role: Video Script Generator
 
@@ -213,11 +280,8 @@ Generate a script for a video, depending on the subject of the video.
 
     final_script = ""
     logger.info(f"subject: {video_subject}")
-    logger.debug(f"prompt: \n{prompt}")
-    response = _generate_response(prompt=prompt)
 
-    # Return the generated script
-    if response:
+    def format_response(response):
         # Clean the script
         # Remove asterisks, hashes
         response = response.replace("*", "")
@@ -234,15 +298,30 @@ Generate a script for a video, depending on the subject of the video.
         selected_paragraphs = paragraphs[:paragraph_number]
 
         # Join the selected paragraphs into a single string
-        final_script = "\n\n".join(selected_paragraphs)
+        return "\n\n".join(paragraphs)
 
-        # Print to console the number of paragraphs used
-        # logger.info(f"number of paragraphs used: {len(selected_paragraphs)}")
-    else:
-        logging.error("gpt returned an empty response")
+    for i in range(_max_retries):
+        try:
+            response = _generate_response(prompt=prompt)
+            if response:
+                final_script = format_response(response)
+            else:
+                logging.error("gpt returned an empty response")
+
+            # g4f may return an error message
+            if final_script and "当日额度已消耗完" in final_script:
+                raise ValueError(final_script)
+
+            if final_script:
+                break
+        except Exception as e:
+            logger.error(f"failed to generate script: {e}")
+
+        if i < _max_retries:
+            logger.warning(f"failed to generate video script, trying again... {i + 1}")
 
     logger.success(f"completed: \n{final_script}")
-    return final_script
+    return final_script.strip()
 
 
 def generate_terms(video_subject: str, video_script: str, amount: int = 5) -> List[str]:
@@ -273,25 +352,34 @@ Please note that you must use English for generating video search terms; Chinese
 """.strip()
 
     logger.info(f"subject: {video_subject}")
-    logger.debug(f"prompt: \n{prompt}")
-    response = _generate_response(prompt)
+
     search_terms = []
+    response = ""
+    for i in range(_max_retries):
+        try:
+            response = _generate_response(prompt)
+            search_terms = json.loads(response)
+            if not isinstance(search_terms, list) or not all(
+                isinstance(term, str) for term in search_terms
+            ):
+                logger.error("response is not a list of strings.")
+                continue
 
-    try:
-        search_terms = json.loads(response)
-        if not isinstance(search_terms, list) or not all(isinstance(term, str) for term in search_terms):
-            raise ValueError("response is not a list of strings.")
+        except Exception as e:
+            logger.warning(f"failed to generate video terms: {str(e)}")
+            if response:
+                match = re.search(r"\[.*]", response)
+                if match:
+                    try:
+                        search_terms = json.loads(match.group())
+                    except Exception as e:
+                        logger.warning(f"failed to generate video terms: {str(e)}")
+                        pass
 
-    except (json.JSONDecodeError, ValueError):
-        # logger.warning(f"gpt returned an unformatted response. attempting to clean...")
-        # Attempt to extract list-like string and convert to list
-        match = re.search(r'\["(?:[^"\\]|\\.)*"(?:,\s*"[^"\\]*")*\]', response)
-        if match:
-            try:
-                search_terms = json.loads(match.group())
-            except json.JSONDecodeError:
-                logger.error(f"could not parse response: {response}")
-                return []
+        if search_terms and len(search_terms) > 0:
+            break
+        if i < _max_retries:
+            logger.warning(f"failed to generate video terms, trying again... {i + 1}")
 
     logger.success(f"completed: \n{search_terms}")
     return search_terms
@@ -299,9 +387,13 @@ Please note that you must use English for generating video search terms; Chinese
 
 if __name__ == "__main__":
     video_subject = "生命的意义是什么"
-    script = generate_script(video_subject=video_subject, language="zh-CN", paragraph_number=1)
-    # print("######################")
-    # print(script)
-    # search_terms = generate_terms(video_subject=video_subject, video_script=script, amount=5)
-    # print("######################")
-    # print(search_terms)
+    script = generate_script(
+        video_subject=video_subject, language="zh-CN", paragraph_number=1
+    )
+    print("######################")
+    print(script)
+    search_terms = generate_terms(
+        video_subject=video_subject, video_script=script, amount=5
+    )
+    print("######################")
+    print(search_terms)

app/services/material.py

@@ -14,44 +14,46 @@ from app.utils import utils
 requested_count = 0
 
 
-def round_robin_api_key():
-    pexels_api_keys = config.app.get("pexels_api_keys")
-    if not pexels_api_keys:
+def get_api_key(cfg_key: str):
+    api_keys = config.app.get(cfg_key)
+    if not api_keys:
         raise ValueError(
-            f"\n\n##### pexels_api_keys is not set #####\n\nPlease set it in the config.toml file: {config.config_file}\n\n{utils.to_json(config.app)}")
+            f"\n\n##### {cfg_key} is not set #####\n\nPlease set it in the config.toml file: {config.config_file}\n\n"
+            f"{utils.to_json(config.app)}"
+        )
 
     # if only one key is provided, return it
-    if isinstance(pexels_api_keys, str):
-        return pexels_api_keys
+    if isinstance(api_keys, str):
+        return api_keys
 
     global requested_count
     requested_count += 1
-    return pexels_api_keys[requested_count % len(pexels_api_keys)]
+    return api_keys[requested_count % len(api_keys)]
 
 
-def search_videos(search_term: str,
-                  minimum_duration: int,
-                  video_aspect: VideoAspect = VideoAspect.portrait,
-                  ) -> List[MaterialInfo]:
+def search_videos_pexels(
+    search_term: str,
+    minimum_duration: int,
+    video_aspect: VideoAspect = VideoAspect.portrait,
+) -> List[MaterialInfo]:
     aspect = VideoAspect(video_aspect)
     video_orientation = aspect.name
     video_width, video_height = aspect.to_resolution()
-
-    headers = {
-        "Authorization": round_robin_api_key()
-    }
-    proxies = config.pexels.get("proxies", None)
+    api_key = get_api_key("pexels_api_keys")
+    headers = {"Authorization": api_key}
     # Build URL
-    params = {
-        "query": search_term,
-        "per_page": 20,
-        "orientation": video_orientation
-    }
+    params = {"query": search_term, "per_page": 20, "orientation": video_orientation}
     query_url = f"https://api.pexels.com/videos/search?{urlencode(params)}"
-    logger.info(f"searching videos: {query_url}, with proxies: {proxies}")
+    logger.info(f"searching videos: {query_url}, with proxies: {config.proxy}")
 
     try:
-        r = requests.get(query_url, headers=headers, proxies=proxies, verify=False)
+        r = requests.get(
+            query_url,
+            headers=headers,
+            proxies=config.proxy,
+            verify=False,
+            timeout=(30, 60),
+        )
         response = r.json()
         video_items = []
         if "videos" not in response:
@@ -83,6 +85,62 @@ def search_videos(search_term: str,
     return []
 
 
+def search_videos_pixabay(
+    search_term: str,
+    minimum_duration: int,
+    video_aspect: VideoAspect = VideoAspect.portrait,
+) -> List[MaterialInfo]:
+    aspect = VideoAspect(video_aspect)
+
+    video_width, video_height = aspect.to_resolution()
+
+    api_key = get_api_key("pixabay_api_keys")
+    # Build URL
+    params = {
+        "q": search_term,
+        "video_type": "all",  # Accepted values: "all", "film", "animation"
+        "per_page": 50,
+        "key": api_key,
+    }
+    query_url = f"https://pixabay.com/api/videos/?{urlencode(params)}"
+    logger.info(f"searching videos: {query_url}, with proxies: {config.proxy}")
+
+    try:
+        r = requests.get(
+            query_url, proxies=config.proxy, verify=False, timeout=(30, 60)
+        )
+        response = r.json()
+        video_items = []
+        if "hits" not in response:
+            logger.error(f"search videos failed: {response}")
+            return video_items
+        videos = response["hits"]
+        # loop through each video in the result
+        for v in videos:
+            duration = v["duration"]
+            # check if video has desired minimum duration
+            if duration < minimum_duration:
+                continue
+            video_files = v["videos"]
+            # loop through each url to determine the best quality
+            for video_type in video_files:
+                video = video_files[video_type]
+                w = int(video["width"])
+                h = int(video["height"])
+                if w >= video_width:
+                    item = MaterialInfo()
+                    item.provider = "pixabay"
+                    item.url = video["url"]
+                    item.duration = duration
+                    video_items.append(item)
+                    break
+        return video_items
+    except Exception as e:
+        logger.error(f"search videos failed: {str(e)}")
+
+    return []
+
+
 def save_video(video_url: str, save_dir: str = "") -> str:
     if not save_dir:
         save_dir = utils.storage_dir("cache_videos")
@@ -101,9 +159,12 @@ def save_video(video_url: str, save_dir: str = "") -> str:
         return video_path
 
     # if video does not exist, download it
-    proxies = config.pexels.get("proxies", None)
     with open(video_path, "wb") as f:
-        f.write(requests.get(video_url, proxies=proxies, verify=False, timeout=(60, 240)).content)
+        f.write(
+            requests.get(
+                video_url, proxies=config.proxy, verify=False, timeout=(60, 240)
+            ).content
+        )
 
     if os.path.exists(video_path) and os.path.getsize(video_path) > 0:
         try:
@@ -122,21 +183,28 @@ def save_video(video_url: str, save_dir: str = "") -> str:
     return ""
 
 
-def download_videos(task_id: str,
-                    search_terms: List[str],
-                    video_aspect: VideoAspect = VideoAspect.portrait,
-                    video_contact_mode: VideoConcatMode = VideoConcatMode.random,
-                    audio_duration: float = 0.0,
-                    max_clip_duration: int = 5,
-                    ) -> List[str]:
+def download_videos(
+    task_id: str,
+    search_terms: List[str],
+    source: str = "pexels",
+    video_aspect: VideoAspect = VideoAspect.portrait,
+    video_contact_mode: VideoConcatMode = VideoConcatMode.random,
+    audio_duration: float = 0.0,
+    max_clip_duration: int = 5,
+) -> List[str]:
     valid_video_items = []
     valid_video_urls = []
     found_duration = 0.0
+    search_videos = search_videos_pexels
+    if source == "pixabay":
+        search_videos = search_videos_pixabay
+
     for search_term in search_terms:
-        # logger.info(f"searching videos for '{search_term}'")
-        video_items = search_videos(search_term=search_term,
-                                    minimum_duration=max_clip_duration,
-                                    video_aspect=video_aspect)
+        video_items = search_videos(
+            search_term=search_term,
+            minimum_duration=max_clip_duration,
+            video_aspect=video_aspect,
+        )
         logger.info(f"found {len(video_items)} videos for '{search_term}'")
 
         for item in video_items:
@@ -146,7 +214,8 @@ def download_videos(task_id: str,
                 found_duration += item.duration
 
     logger.info(
-        f"found total videos: {len(valid_video_items)}, required duration: {audio_duration} seconds, found duration: {found_duration} seconds")
+        f"found total videos: {len(valid_video_items)}, required duration: {audio_duration} seconds, found duration: {found_duration} seconds"
+    )
     video_paths = []
 
     material_directory = config.app.get("material_directory", "").strip()
@@ -162,14 +231,18 @@ def download_videos(task_id: str,
     for item in valid_video_items:
         try:
             logger.info(f"downloading video: {item.url}")
-            saved_video_path = save_video(video_url=item.url, save_dir=material_directory)
+            saved_video_path = save_video(
+                video_url=item.url, save_dir=material_directory
+            )
             if saved_video_path:
                 logger.info(f"video saved: {saved_video_path}")
                 video_paths.append(saved_video_path)
                 seconds = min(max_clip_duration, item.duration)
                 total_duration += seconds
                 if total_duration > audio_duration:
-                    logger.info(f"total duration of downloaded videos: {total_duration} seconds, skip downloading more")
+                    logger.info(
+                        f"total duration of downloaded videos: {total_duration} seconds, skip downloading more"
+                    )
                     break
         except Exception as e:
             logger.error(f"failed to download video: {utils.to_json(item)} => {str(e)}")
@@ -178,4 +251,6 @@ def download_videos(task_id: str,
 
 
 if __name__ == "__main__":
-    download_videos("test123", ["cat"], audio_duration=100)
+    download_videos(
+        "test123", ["Money Exchange Medium"], audio_duration=100, source="pixabay"
+    )

app/services/state.py

@@ -6,7 +6,6 @@ from app.models import const
 
 # Base class for state management
 class BaseState(ABC):
-
     @abstractmethod
     def update_task(self, task_id: str, state: int, progress: int = 0, **kwargs):
         pass
@@ -18,11 +17,16 @@ class BaseState(ABC):
 
 # Memory state management
 class MemoryState(BaseState):
-
     def __init__(self):
         self._tasks = {}
 
-    def update_task(self, task_id: str, state: int = const.TASK_STATE_PROCESSING, progress: int = 0, **kwargs):
+    def update_task(
+        self,
+        task_id: str,
+        state: int = const.TASK_STATE_PROCESSING,
+        progress: int = 0,
+        **kwargs,
+    ):
         progress = int(progress)
         if progress > 100:
             progress = 100
@@ -43,12 +47,18 @@ class MemoryState(BaseState):
 
 # Redis state management
 class RedisState(BaseState):
-
-    def __init__(self, host='localhost', port=6379, db=0, password=None):
+    def __init__(self, host="localhost", port=6379, db=0, password=None):
         import redis
+
         self._redis = redis.StrictRedis(host=host, port=port, db=db, password=password)
 
-    def update_task(self, task_id: str, state: int = const.TASK_STATE_PROCESSING, progress: int = 0, **kwargs):
+    def update_task(
+        self,
+        task_id: str,
+        state: int = const.TASK_STATE_PROCESSING,
+        progress: int = 0,
+        **kwargs,
+    ):
         progress = int(progress)
         if progress > 100:
             progress = 100
@@ -67,7 +77,10 @@ class RedisState(BaseState):
         if not task_data:
             return None
 
-        task = {key.decode('utf-8'): self._convert_to_original_type(value) for key, value in task_data.items()}
+        task = {
+            key.decode("utf-8"): self._convert_to_original_type(value)
+            for key, value in task_data.items()
+        }
         return task
 
     def delete_task(self, task_id: str):
@@ -79,7 +92,7 @@ class RedisState(BaseState):
         Convert the value from byte string to its original data type.
         You can extend this method to handle other data types as needed.
         """
-        value_str = value.decode('utf-8')
+        value_str = value.decode("utf-8")
 
         try:
             # try to convert byte string array to list
@@ -100,4 +113,10 @@ _redis_port = config.app.get("redis_port", 6379)
 _redis_db = config.app.get("redis_db", 0)
 _redis_password = config.app.get("redis_password", None)
 
-state = RedisState(host=_redis_host, port=_redis_port, db=_redis_db, password=_redis_password) if _enable_redis else MemoryState()
+state = (
+    RedisState(
+        host=_redis_host, port=_redis_port, db=_redis_db, password=_redis_password
+    )
+    if _enable_redis
+    else MemoryState()
+)

app/services/subtitle.py

@@ -23,10 +23,23 @@ def create(audio_file, subtitle_file: str = ""):
         if not os.path.isdir(model_path) or not os.path.isfile(model_bin_file):
             model_path = model_size
 
-        logger.info(f"loading model: {model_path}, device: {device}, compute_type: {compute_type}")
-        model = WhisperModel(model_size_or_path=model_path,
-                             device=device,
-                             compute_type=compute_type)
+        logger.info(
+            f"loading model: {model_path}, device: {device}, compute_type: {compute_type}"
+        )
+        try:
+            model = WhisperModel(
+                model_size_or_path=model_path, device=device, compute_type=compute_type
+            )
+        except Exception as e:
+            logger.error(
+                f"failed to load model: {e} \n\n"
+                f"********************************************\n"
+                f"this may be caused by network issue. \n"
+                f"please download the model manually and put it in the 'models' folder. \n"
+                f"see [README.md FAQ](https://github.com/harry0703/MoneyPrinterTurbo) for more details.\n"
+                f"********************************************\n\n"
+            )
+            return None
 
     logger.info(f"start, output file: {subtitle_file}")
     if not subtitle_file:
@@ -40,7 +53,9 @@ def create(audio_file, subtitle_file: str = ""):
         vad_parameters=dict(min_silence_duration_ms=500),
     )
 
-    logger.info(f"detected language: '{info.language}', probability: {info.language_probability:.2f}")
+    logger.info(
+        f"detected language: '{info.language}', probability: {info.language_probability:.2f}"
+    )
 
     start = timer()
     subtitles = []
@@ -53,11 +68,9 @@ def create(audio_file, subtitle_file: str = ""):
         msg = "[%.2fs -> %.2fs] %s" % (seg_start, seg_end, seg_text)
         logger.debug(msg)
 
-        subtitles.append({
-            "msg": seg_text,
-            "start_time": seg_start,
-            "end_time": seg_end
-        })
+        subtitles.append(
+            {"msg": seg_text, "start_time": seg_start, "end_time": seg_end}
+        )
 
     for segment in segments:
         words_idx = 0
@@ -110,7 +123,11 @@ def create(audio_file, subtitle_file: str = ""):
     for subtitle in subtitles:
         text = subtitle.get("msg")
         if text:
-            lines.append(utils.text_to_srt(idx, text, subtitle.get("start_time"), subtitle.get("end_time")))
+            lines.append(
+                utils.text_to_srt(
+                    idx, text, subtitle.get("start_time"), subtitle.get("end_time")
+                )
+            )
             idx += 1
 
     sub = "\n".join(lines) + "\n"
@@ -120,16 +137,19 @@ def create(audio_file, subtitle_file: str = ""):
 
 
 def file_to_subtitles(filename):
+    if not filename or not os.path.isfile(filename):
+        return []
+
     times_texts = []
     current_times = None
     current_text = ""
     index = 0
-    with open(filename, 'r', encoding="utf-8") as f:
+    with open(filename, "r", encoding="utf-8") as f:
         for line in f:
             times = re.findall("([0-9]*:[0-9]*:[0-9]*,[0-9]*)", line)
             if times:
                 current_times = line
-            elif line.strip() == '' and current_times:
+            elif line.strip() == "" and current_times:
                 index += 1
                 times_texts.append((index, current_times.strip(), current_text.strip()))
                 current_times, current_text = None, ""
@@ -138,27 +158,124 @@ def file_to_subtitles(filename):
     return times_texts
 
 
+def levenshtein_distance(s1, s2):
+    if len(s1) < len(s2):
+        return levenshtein_distance(s2, s1)
+
+    if len(s2) == 0:
+        return len(s1)
+
+    previous_row = range(len(s2) + 1)
+    for i, c1 in enumerate(s1):
+        current_row = [i + 1]
+        for j, c2 in enumerate(s2):
+            insertions = previous_row[j + 1] + 1
+            deletions = current_row[j] + 1
+            substitutions = previous_row[j] + (c1 != c2)
+            current_row.append(min(insertions, deletions, substitutions))
+        previous_row = current_row
+
+    return previous_row[-1]
+
+
+def similarity(a, b):
+    distance = levenshtein_distance(a.lower(), b.lower())
+    max_length = max(len(a), len(b))
+    return 1 - (distance / max_length)
+
+
 def correct(subtitle_file, video_script):
     subtitle_items = file_to_subtitles(subtitle_file)
     script_lines = utils.split_string_by_punctuations(video_script)
 
     corrected = False
-    if len(subtitle_items) == len(script_lines):
-        for i in range(len(script_lines)):
-            script_line = script_lines[i].strip()
-            subtitle_line = subtitle_items[i][2]
-            if script_line != subtitle_line:
-                logger.warning(f"line {i + 1}, script: {script_line}, subtitle: {subtitle_line}")
-                subtitle_items[i] = (subtitle_items[i][0], subtitle_items[i][1], script_line)
+    new_subtitle_items = []
+    script_index = 0
+    subtitle_index = 0
+
+    while script_index < len(script_lines) and subtitle_index < len(subtitle_items):
+        script_line = script_lines[script_index].strip()
+        subtitle_line = subtitle_items[subtitle_index][2].strip()
+
+        if script_line == subtitle_line:
+            new_subtitle_items.append(subtitle_items[subtitle_index])
+            script_index += 1
+            subtitle_index += 1
+        else:
+            combined_subtitle = subtitle_line
+            start_time = subtitle_items[subtitle_index][1].split(" --> ")[0]
+            end_time = subtitle_items[subtitle_index][1].split(" --> ")[1]
+            next_subtitle_index = subtitle_index + 1
+
+            while next_subtitle_index < len(subtitle_items):
+                next_subtitle = subtitle_items[next_subtitle_index][2].strip()
+                if similarity(
+                    script_line, combined_subtitle + " " + next_subtitle
+                ) > similarity(script_line, combined_subtitle):
+                    combined_subtitle += " " + next_subtitle
+                    end_time = subtitle_items[next_subtitle_index][1].split(" --> ")[1]
+                    next_subtitle_index += 1
+                else:
+                    break
+
+            if similarity(script_line, combined_subtitle) > 0.8:
+                logger.warning(
+                    f"Merged/Corrected - Script: {script_line}, Subtitle: {combined_subtitle}"
+                )
+                new_subtitle_items.append(
+                    (
+                        len(new_subtitle_items) + 1,
+                        f"{start_time} --> {end_time}",
+                        script_line,
+                    )
+                )
                 corrected = True
+            else:
+                logger.warning(
+                    f"Mismatch - Script: {script_line}, Subtitle: {combined_subtitle}"
+                )
+                new_subtitle_items.append(
+                    (
+                        len(new_subtitle_items) + 1,
+                        f"{start_time} --> {end_time}",
+                        script_line,
+                    )
+                )
+                corrected = True
+
+            script_index += 1
+            subtitle_index = next_subtitle_index
+
+    # 处理剩余的脚本行
+    while script_index < len(script_lines):
+        logger.warning(f"Extra script line: {script_lines[script_index]}")
+        if subtitle_index < len(subtitle_items):
+            new_subtitle_items.append(
+                (
+                    len(new_subtitle_items) + 1,
+                    subtitle_items[subtitle_index][1],
+                    script_lines[script_index],
+                )
+            )
+            subtitle_index += 1
+        else:
+            new_subtitle_items.append(
+                (
+                    len(new_subtitle_items) + 1,
+                    "00:00:00,000 --> 00:00:00,000",
+                    script_lines[script_index],
+                )
+            )
+        script_index += 1
+        corrected = True
 
     if corrected:
         with open(subtitle_file, "w", encoding="utf-8") as fd:
-            for item in subtitle_items:
-                fd.write(f"{item[0]}\n{item[1]}\n{item[2]}\n\n")
-        logger.info(f"subtitle corrected")
+            for i, item in enumerate(new_subtitle_items):
+                fd.write(f"{i + 1}\n{item[1]}\n{item[2]}\n\n")
+        logger.info("Subtitle corrected")
     else:
-        logger.success(f"subtitle is correct")
+        logger.success("Subtitle is correct")
 
 
 if __name__ == "__main__":

app/services/task.py

@@ -7,52 +7,42 @@ from loguru import logger
 
 from app.config import config
 from app.models import const
-from app.models.schema import VideoParams, VideoConcatMode
-from app.services import llm, material, voice, video, subtitle
+from app.models.schema import VideoConcatMode, VideoParams
+from app.services import llm, material, subtitle, video, voice
 from app.services import state as sm
 from app.utils import utils
 
 
-def start(task_id, params: VideoParams):
-    """
-    {
-        "video_subject": "",
-        "video_aspect": "横屏 16:9（西瓜视频）",
-        "voice_name": "女生-晓晓",
-        "enable_bgm": false,
-        "font_name": "STHeitiMedium 黑体-中",
-        "text_color": "#FFFFFF",
-        "font_size": 60,
-        "stroke_color": "#000000",
-        "stroke_width": 1.5
-    }
-    """
-    logger.info(f"start task: {task_id}")
-    sm.state.update_task(task_id, state=const.TASK_STATE_PROCESSING, progress=5)
-
-    video_subject = params.video_subject
-    voice_name = voice.parse_voice_name(params.voice_name)
-    paragraph_number = params.paragraph_number
-    n_threads = params.n_threads
-    max_clip_duration = params.video_clip_duration
-
+def generate_script(task_id, params):
     logger.info("\n\n## generating video script")
     video_script = params.video_script.strip()
     if not video_script:
-        video_script = llm.generate_script(video_subject=video_subject, language=params.video_language,
-                                           paragraph_number=paragraph_number)
+        video_script = llm.generate_script(
+            video_subject=params.video_subject,
+            language=params.video_language,
+            paragraph_number=params.paragraph_number,
+        )
     else:
         logger.debug(f"video script: \n{video_script}")
 
-    sm.state.update_task(task_id, state=const.TASK_STATE_PROCESSING, progress=10)
+    if not video_script:
+        sm.state.update_task(task_id, state=const.TASK_STATE_FAILED)
+        logger.error("failed to generate video script.")
+        return None
 
+    return video_script
+
+
+def generate_terms(task_id, params, video_script):
     logger.info("\n\n## generating video terms")
     video_terms = params.video_terms
     if not video_terms:
-        video_terms = llm.generate_terms(video_subject=video_subject, video_script=video_script, amount=5)
+        video_terms = llm.generate_terms(
+            video_subject=params.video_subject, video_script=video_script, amount=5
+        )
     else:
         if isinstance(video_terms, str):
-            video_terms = [term.strip() for term in re.split(r'[,，]', video_terms)]
+            video_terms = [term.strip() for term in re.split(r"[,，]", video_terms)]
         elif isinstance(video_terms, list):
             video_terms = [term.strip() for term in video_terms]
         else:
@@ -60,7 +50,16 @@ def start(task_id, params: VideoParams):
 
         logger.debug(f"video terms: {utils.to_json(video_terms)}")
 
-    script_file = path.join(utils.task_dir(task_id), f"script.json")
+    if not video_terms:
+        sm.state.update_task(task_id, state=const.TASK_STATE_FAILED)
+        logger.error("failed to generate video terms.")
+        return None
+
+    return video_terms
+
+
+def save_script_data(task_id, video_script, video_terms, params):
+    script_file = path.join(utils.task_dir(task_id), "script.json")
     script_data = {
         "script": video_script,
         "search_terms": video_terms,
@@ -70,80 +69,118 @@ def start(task_id, params: VideoParams):
     with open(script_file, "w", encoding="utf-8") as f:
         f.write(utils.to_json(script_data))
 
-    sm.state.update_task(task_id, state=const.TASK_STATE_PROCESSING, progress=20)
 
+def generate_audio(task_id, params, video_script):
     logger.info("\n\n## generating audio")
-    audio_file = path.join(utils.task_dir(task_id), f"audio.mp3")
-    sub_maker = voice.tts(text=video_script, voice_name=voice_name, voice_file=audio_file)
+    audio_file = path.join(utils.task_dir(task_id), "audio.mp3")
+    sub_maker = voice.tts(
+        text=video_script,
+        voice_name=voice.parse_voice_name(params.voice_name),
+        voice_rate=params.voice_rate,
+        voice_file=audio_file,
+    )
     if sub_maker is None:
         sm.state.update_task(task_id, state=const.TASK_STATE_FAILED)
         logger.error(
-            "failed to generate audio, maybe the network is not available. if you are in China, please use a VPN.")
-        return
+            """failed to generate audio:
+1. check if the language of the voice matches the language of the video script.
+2. check if the network is available. If you are in China, it is recommended to use a VPN and enable the global traffic mode.
+        """.strip()
+        )
+        return None, None
 
-    audio_duration = voice.get_audio_duration(sub_maker)
-    audio_duration = math.ceil(audio_duration)
+    audio_duration = math.ceil(voice.get_audio_duration(sub_maker))
+    return audio_file, audio_duration
 
-    sm.state.update_task(task_id, state=const.TASK_STATE_PROCESSING, progress=30)
 
-    subtitle_path = ""
-    if params.subtitle_enabled:
-        subtitle_path = path.join(utils.task_dir(task_id), f"subtitle.srt")
-        subtitle_provider = config.app.get("subtitle_provider", "").strip().lower()
-        logger.info(f"\n\n## generating subtitle, provider: {subtitle_provider}")
-        subtitle_fallback = False
-        if subtitle_provider == "edge":
-            voice.create_subtitle(text=video_script, sub_maker=sub_maker, subtitle_file=subtitle_path)
-            if not os.path.exists(subtitle_path):
-                subtitle_fallback = True
-                logger.warning("subtitle file not found, fallback to whisper")
+def generate_subtitle(task_id, params, video_script, sub_maker, audio_file):
+    if not params.subtitle_enabled:
+        return ""
 
-        if subtitle_provider == "whisper" or subtitle_fallback:
-            subtitle.create(audio_file=audio_file, subtitle_file=subtitle_path)
-            logger.info("\n\n## correcting subtitle")
-            subtitle.correct(subtitle_file=subtitle_path, video_script=video_script)
+    subtitle_path = path.join(utils.task_dir(task_id), "subtitle.srt")
+    subtitle_provider = config.app.get("subtitle_provider", "").strip().lower()
+    logger.info(f"\n\n## generating subtitle, provider: {subtitle_provider}")
 
-        subtitle_lines = subtitle.file_to_subtitles(subtitle_path)
-        if not subtitle_lines:
-            logger.warning(f"subtitle file is invalid: {subtitle_path}")
-            subtitle_path = ""
+    subtitle_fallback = False
+    if subtitle_provider == "edge":
+        voice.create_subtitle(
+            text=video_script, sub_maker=sub_maker, subtitle_file=subtitle_path
+        )
+        if not os.path.exists(subtitle_path):
+            subtitle_fallback = True
+            logger.warning("subtitle file not found, fallback to whisper")
 
-    sm.state.update_task(task_id, state=const.TASK_STATE_PROCESSING, progress=40)
+    if subtitle_provider == "whisper" or subtitle_fallback:
+        subtitle.create(audio_file=audio_file, subtitle_file=subtitle_path)
+        logger.info("\n\n## correcting subtitle")
+        subtitle.correct(subtitle_file=subtitle_path, video_script=video_script)
 
-    logger.info("\n\n## downloading videos")
-    downloaded_videos = material.download_videos(task_id=task_id,
-                                                 search_terms=video_terms,
-                                                 video_aspect=params.video_aspect,
-                                                 video_contact_mode=params.video_concat_mode,
-                                                 audio_duration=audio_duration * params.video_count,
-                                                 max_clip_duration=max_clip_duration,
-                                                 )
-    if not downloaded_videos:
-        sm.state.update_task(task_id, state=const.TASK_STATE_FAILED)
-        logger.error(
-            "failed to download videos, maybe the network is not available. if you are in China, please use a VPN.")
-        return
+    subtitle_lines = subtitle.file_to_subtitles(subtitle_path)
+    if not subtitle_lines:
+        logger.warning(f"subtitle file is invalid: {subtitle_path}")
+        return ""
 
-    sm.state.update_task(task_id, state=const.TASK_STATE_PROCESSING, progress=50)
+    return subtitle_path
 
+
+def get_video_materials(task_id, params, video_terms, audio_duration):
+    if params.video_source == "local":
+        logger.info("\n\n## preprocess local materials")
+        materials = video.preprocess_video(
+            materials=params.video_materials, clip_duration=params.video_clip_duration
+        )
+        if not materials:
+            sm.state.update_task(task_id, state=const.TASK_STATE_FAILED)
+            logger.error(
+                "no valid materials found, please check the materials and try again."
+            )
+            return None
+        return [material_info.url for material_info in materials]
+    else:
+        logger.info(f"\n\n## downloading videos from {params.video_source}")
+        downloaded_videos = material.download_videos(
+            task_id=task_id,
+            search_terms=video_terms,
+            source=params.video_source,
+            video_aspect=params.video_aspect,
+            video_contact_mode=params.video_concat_mode,
+            audio_duration=audio_duration * params.video_count,
+            max_clip_duration=params.video_clip_duration,
+        )
+        if not downloaded_videos:
+            sm.state.update_task(task_id, state=const.TASK_STATE_FAILED)
+            logger.error(
+                "failed to download videos, maybe the network is not available. if you are in China, please use a VPN."
+            )
+            return None
+        return downloaded_videos
+
+
+def generate_final_videos(
+    task_id, params, downloaded_videos, audio_file, subtitle_path
+):
     final_video_paths = []
     combined_video_paths = []
-    video_concat_mode = params.video_concat_mode
-    if params.video_count > 1:
-        video_concat_mode = VideoConcatMode.random
+    video_concat_mode = (
+        params.video_concat_mode if params.video_count == 1 else VideoConcatMode.random
+    )
 
     _progress = 50
     for i in range(params.video_count):
         index = i + 1
-        combined_video_path = path.join(utils.task_dir(task_id), f"combined-{index}.mp4")
+        combined_video_path = path.join(
+            utils.task_dir(task_id), f"combined-{index}.mp4"
+        )
         logger.info(f"\n\n## combining video: {index} => {combined_video_path}")
-        video.combine_videos(combined_video_path=combined_video_path,
-                             video_paths=downloaded_videos,
-                             audio_file=audio_file,
-                             video_aspect=params.video_aspect,
-                             video_concat_mode=video_concat_mode,
-                             max_clip_duration=max_clip_duration,
-                             threads=n_threads)
+        video.combine_videos(
+            combined_video_path=combined_video_path,
+            video_paths=downloaded_videos,
+            audio_file=audio_file,
+            video_aspect=params.video_aspect,
+            video_concat_mode=video_concat_mode,
+            max_clip_duration=params.video_clip_duration,
+            threads=params.n_threads,
+        )
 
         _progress += 50 / params.video_count / 2
         sm.state.update_task(task_id, progress=_progress)
@@ -151,13 +188,13 @@ def start(task_id, params: VideoParams):
         final_video_path = path.join(utils.task_dir(task_id), f"final-{index}.mp4")
 
         logger.info(f"\n\n## generating video: {index} => {final_video_path}")
-        # Put everything together
-        video.generate_video(video_path=combined_video_path,
-                             audio_path=audio_file,
-                             subtitle_path=subtitle_path,
-                             output_file=final_video_path,
-                             params=params,
-                             )
+        video.generate_video(
+            video_path=combined_video_path,
+            audio_path=audio_file,
+            subtitle_path=subtitle_path,
+            output_file=final_video_path,
+            params=params,
+        )
 
         _progress += 50 / params.video_count / 2
         sm.state.update_task(task_id, progress=_progress)
@@ -165,11 +202,119 @@ def start(task_id, params: VideoParams):
         final_video_paths.append(final_video_path)
         combined_video_paths.append(combined_video_path)
 
-    logger.success(f"task {task_id} finished, generated {len(final_video_paths)} videos.")
+    return final_video_paths, combined_video_paths
+
+
+def start(task_id, params: VideoParams, stop_at: str = "video"):
+    logger.info(f"start task: {task_id}, stop_at: {stop_at}")
+    sm.state.update_task(task_id, state=const.TASK_STATE_PROCESSING, progress=5)
+
+    # 1. Generate script
+    video_script = generate_script(task_id, params)
+    if not video_script:
+        sm.state.update_task(task_id, state=const.TASK_STATE_FAILED)
+        return
+
+    sm.state.update_task(task_id, state=const.TASK_STATE_PROCESSING, progress=10)
+
+    if stop_at == "script":
+        sm.state.update_task(
+            task_id, state=const.TASK_STATE_COMPLETE, progress=100, script=video_script
+        )
+        return {"script": video_script}
+
+    # 2. Generate terms
+    video_terms = ""
+    if params.video_source != "local":
+        video_terms = generate_terms(task_id, params, video_script)
+        if not video_terms:
+            sm.state.update_task(task_id, state=const.TASK_STATE_FAILED)
+            return
+
+    save_script_data(task_id, video_script, video_terms, params)
+
+    if stop_at == "terms":
+        sm.state.update_task(
+            task_id, state=const.TASK_STATE_COMPLETE, progress=100, terms=video_terms
+        )
+        return {"script": video_script, "terms": video_terms}
+
+    sm.state.update_task(task_id, state=const.TASK_STATE_PROCESSING, progress=20)
+
+    # 3. Generate audio
+    audio_file, audio_duration = generate_audio(task_id, params, video_script)
+    if not audio_file:
+        sm.state.update_task(task_id, state=const.TASK_STATE_FAILED)
+        return
+
+    sm.state.update_task(task_id, state=const.TASK_STATE_PROCESSING, progress=30)
+
+    if stop_at == "audio":
+        sm.state.update_task(
+            task_id,
+            state=const.TASK_STATE_COMPLETE,
+            progress=100,
+            audio_file=audio_file,
+        )
+        return {"audio_file": audio_file, "audio_duration": audio_duration}
+
+    # 4. Generate subtitle
+    subtitle_path = generate_subtitle(task_id, params, video_script, None, audio_file)
+
+    if stop_at == "subtitle":
+        sm.state.update_task(
+            task_id,
+            state=const.TASK_STATE_COMPLETE,
+            progress=100,
+            subtitle_path=subtitle_path,
+        )
+        return {"subtitle_path": subtitle_path}
+
+    sm.state.update_task(task_id, state=const.TASK_STATE_PROCESSING, progress=40)
+
+    # 5. Get video materials
+    downloaded_videos = get_video_materials(
+        task_id, params, video_terms, audio_duration
+    )
+    if not downloaded_videos:
+        sm.state.update_task(task_id, state=const.TASK_STATE_FAILED)
+        return
+
+    if stop_at == "materials":
+        sm.state.update_task(
+            task_id,
+            state=const.TASK_STATE_COMPLETE,
+            progress=100,
+            materials=downloaded_videos,
+        )
+        return {"materials": downloaded_videos}
+
+    sm.state.update_task(task_id, state=const.TASK_STATE_PROCESSING, progress=50)
+
+    # 6. Generate final videos
+    final_video_paths, combined_video_paths = generate_final_videos(
+        task_id, params, downloaded_videos, audio_file, subtitle_path
+    )
+
+    if not final_video_paths:
+        sm.state.update_task(task_id, state=const.TASK_STATE_FAILED)
+        return
+
+    logger.success(
+        f"task {task_id} finished, generated {len(final_video_paths)} videos."
+    )
 
     kwargs = {
         "videos": final_video_paths,
-        "combined_videos": combined_video_paths
+        "combined_videos": combined_video_paths,
+        "script": video_script,
+        "terms": video_terms,
+        "audio_file": audio_file,
+        "audio_duration": audio_duration,
+        "subtitle_path": subtitle_path,
+        "materials": downloaded_videos,
     }
-    sm.state.update_task(task_id, state=const.TASK_STATE_COMPLETE, progress=100, **kwargs)
+    sm.state.update_task(
+        task_id, state=const.TASK_STATE_COMPLETE, progress=100, **kwargs
+    )
     return kwargs

app/services/video.py

@@ -1,12 +1,14 @@
 import glob
 import random
 from typing import List
-from PIL import ImageFont
+
 from loguru import logger
 from moviepy.editor import *
 from moviepy.video.tools.subtitles import SubtitlesClip
+from PIL import ImageFont
 
-from app.models.schema import VideoAspect, VideoParams, VideoConcatMode
+from app.models import const
+from app.models.schema import MaterialInfo, VideoAspect, VideoConcatMode, VideoParams
 from app.utils import utils
 
 
@@ -26,14 +28,15 @@ def get_bgm_file(bgm_type: str = "random", bgm_file: str = ""):
     return ""
 
 
-def combine_videos(combined_video_path: str,
-                   video_paths: List[str],
-                   audio_file: str,
-                   video_aspect: VideoAspect = VideoAspect.portrait,
-                   video_concat_mode: VideoConcatMode = VideoConcatMode.random,
-                   max_clip_duration: int = 5,
-                   threads: int = 2,
-                   ) -> str:
+def combine_videos(
+    combined_video_path: str,
+    video_paths: List[str],
+    audio_file: str,
+    video_aspect: VideoAspect = VideoAspect.portrait,
+    video_concat_mode: VideoConcatMode = VideoConcatMode.random,
+    max_clip_duration: int = 5,
+    threads: int = 2,
+) -> str:
     audio_clip = AudioFileClip(audio_file)
     audio_duration = audio_clip.duration
     logger.info(f"max duration of audio: {audio_duration} seconds")
@@ -48,14 +51,29 @@ def combine_videos(combined_video_path: str,
 
     clips = []
     video_duration = 0
+
+    raw_clips = []
+    for video_path in video_paths:
+        clip = VideoFileClip(video_path).without_audio()
+        clip_duration = clip.duration
+        start_time = 0
+
+        while start_time < clip_duration:
+            end_time = min(start_time + max_clip_duration, clip_duration)
+            split_clip = clip.subclip(start_time, end_time)
+            raw_clips.append(split_clip)
+            # logger.info(f"splitting from {start_time:.2f} to {end_time:.2f}, clip duration {clip_duration:.2f}, split_clip duration {split_clip.duration:.2f}")
+            start_time = end_time
+            if video_concat_mode.value == VideoConcatMode.sequential.value:
+                break
+
+    # random video_paths order
+    if video_concat_mode.value == VideoConcatMode.random.value:
+        random.shuffle(raw_clips)
+
     # Add downloaded clips over and over until the duration of the audio (max_duration) has been reached
     while video_duration < audio_duration:
-        # random video_paths order
-        if video_concat_mode.value == VideoConcatMode.random.value:
-            random.shuffle(video_paths)
-
-        for video_path in video_paths:
-            clip = VideoFileClip(video_path).without_audio()
+        for clip in raw_clips:
             # Check if clip is longer than the remaining audio
             if (audio_duration - video_duration) < clip.duration:
                 clip = clip.subclip(0, (audio_duration - video_duration))
@@ -86,13 +104,19 @@ def combine_videos(combined_video_path: str,
                     new_height = int(clip_h * scale_factor)
                     clip_resized = clip.resize(newsize=(new_width, new_height))
 
-                    background = ColorClip(size=(video_width, video_height), color=(0, 0, 0))
-                    clip = CompositeVideoClip([
-                        background.set_duration(clip.duration),
-                        clip_resized.set_position("center")
-                    ])
+                    background = ColorClip(
+                        size=(video_width, video_height), color=(0, 0, 0)
+                    )
+                    clip = CompositeVideoClip(
+                        [
+                            background.set_duration(clip.duration),
+                            clip_resized.set_position("center"),
+                        ]
+                    )
 
-                logger.info(f"resizing video to {video_width} x {video_height}, clip size: {clip_w} x {clip_h}")
+                logger.info(
+                    f"resizing video to {video_width} x {video_height}, clip size: {clip_w} x {clip_h}"
+                )
 
             if clip.duration > max_clip_duration:
                 clip = clip.subclip(0, max_clip_duration)
@@ -102,21 +126,22 @@ def combine_videos(combined_video_path: str,
 
     video_clip = concatenate_videoclips(clips)
     video_clip = video_clip.set_fps(30)
-    logger.info(f"writing")
+    logger.info("writing")
     # https://github.com/harry0703/MoneyPrinterTurbo/issues/111#issuecomment-2032354030
-    video_clip.write_videofile(filename=combined_video_path,
-                               threads=threads,
-                               logger=None,
-                               temp_audiofile_path=output_dir,
-                               audio_codec="aac",
-                               fps=30,
-                               )
+    video_clip.write_videofile(
+        filename=combined_video_path,
+        threads=threads,
+        logger=None,
+        temp_audiofile_path=output_dir,
+        audio_codec="aac",
+        fps=30,
+    )
     video_clip.close()
-    logger.success(f"completed")
+    logger.success("completed")
     return combined_video_path
 
 
-def wrap_text(text, max_width, font='Arial', fontsize=60):
+def wrap_text(text, max_width, font="Arial", fontsize=60):
     # 创建字体对象
     font = ImageFont.truetype(font, fontsize)
 
@@ -135,7 +160,7 @@ def wrap_text(text, max_width, font='Arial', fontsize=60):
 
     _wrapped_lines_ = []
     words = text.split(" ")
-    _txt_ = ''
+    _txt_ = ""
     for word in words:
         _before = _txt_
         _txt_ += f"{word} "
@@ -151,14 +176,14 @@ def wrap_text(text, max_width, font='Arial', fontsize=60):
     _wrapped_lines_.append(_txt_)
     if processed:
         _wrapped_lines_ = [line.strip() for line in _wrapped_lines_]
-        result = '\n'.join(_wrapped_lines_).strip()
+        result = "\n".join(_wrapped_lines_).strip()
         height = len(_wrapped_lines_) * height
         # logger.warning(f"wrapped text: {result}")
         return result, height
 
     _wrapped_lines_ = []
     chars = list(text)
-    _txt_ = ''
+    _txt_ = ""
     for word in chars:
         _txt_ += word
         _width, _height = get_text_size(_txt_)
@@ -166,20 +191,21 @@ def wrap_text(text, max_width, font='Arial', fontsize=60):
             continue
         else:
             _wrapped_lines_.append(_txt_)
-            _txt_ = ''
+            _txt_ = ""
     _wrapped_lines_.append(_txt_)
-    result = '\n'.join(_wrapped_lines_).strip()
+    result = "\n".join(_wrapped_lines_).strip()
     height = len(_wrapped_lines_) * height
     # logger.warning(f"wrapped text: {result}")
     return result, height
 
 
-def generate_video(video_path: str,
-                   audio_path: str,
-                   subtitle_path: str,
-                   output_file: str,
-                   params: VideoParams,
-                   ):
+def generate_video(
+    video_path: str,
+    audio_path: str,
+    subtitle_path: str,
+    output_file: str,
+    params: VideoParams,
+):
     aspect = VideoAspect(params.video_aspect)
     video_width, video_height = aspect.to_resolution()
 
@@ -199,7 +225,7 @@ def generate_video(video_path: str,
         if not params.font_name:
             params.font_name = "STHeitiMedium.ttc"
         font_path = os.path.join(utils.font_dir(), params.font_name)
-        if os.name == 'nt':
+        if os.name == "nt":
             font_path = font_path.replace("\\", "/")
 
         logger.info(f"using font: {font_path}")
@@ -207,11 +233,9 @@ def generate_video(video_path: str,
     def create_text_clip(subtitle_item):
         phrase = subtitle_item[1]
         max_width = video_width * 0.9
-        wrapped_txt, txt_height = wrap_text(phrase,
-                                            max_width=max_width,
-                                            font=font_path,
-                                            fontsize=params.font_size
-                                            )
+        wrapped_txt, txt_height = wrap_text(
+            phrase, max_width=max_width, font=font_path, fontsize=params.font_size
+        )
         _clip = TextClip(
             wrapped_txt,
             font=font_path,
@@ -227,18 +251,26 @@ def generate_video(video_path: str,
         _clip = _clip.set_end(subtitle_item[0][1])
         _clip = _clip.set_duration(duration)
         if params.subtitle_position == "bottom":
-            _clip = _clip.set_position(('center', video_height * 0.95 - _clip.h))
+            _clip = _clip.set_position(("center", video_height * 0.95 - _clip.h))
         elif params.subtitle_position == "top":
-            _clip = _clip.set_position(('center', video_height * 0.1))
-        else:
-            _clip = _clip.set_position(('center', 'center'))
+            _clip = _clip.set_position(("center", video_height * 0.05))
+        elif params.subtitle_position == "custom":
+            # 确保字幕完全在屏幕内
+            margin = 10  # 额外的边距，单位为像素
+            max_y = video_height - _clip.h - margin
+            min_y = margin
+            custom_y = (video_height - _clip.h) * (params.custom_position / 100)
+            custom_y = max(min_y, min(custom_y, max_y))  # 限制 y 值在有效范围内
+            _clip = _clip.set_position(("center", custom_y))
+        else:  # center
+            _clip = _clip.set_position(("center", "center"))
         return _clip
 
     video_clip = VideoFileClip(video_path)
     audio_clip = AudioFileClip(audio_path).volumex(params.voice_volume)
 
     if subtitle_path and os.path.exists(subtitle_path):
-        sub = SubtitlesClip(subtitles=subtitle_path, encoding='utf-8')
+        sub = SubtitlesClip(subtitles=subtitle_path, encoding="utf-8")
         text_clips = []
         for item in sub.subtitles:
             clip = create_text_clip(subtitle_item=item)
@@ -248,75 +280,130 @@ def generate_video(video_path: str,
     bgm_file = get_bgm_file(bgm_type=params.bgm_type, bgm_file=params.bgm_file)
     if bgm_file:
         try:
-            bgm_clip = (AudioFileClip(bgm_file)
-                        .volumex(params.bgm_volume)
-                        .audio_fadeout(3))
+            bgm_clip = (
+                AudioFileClip(bgm_file).volumex(params.bgm_volume).audio_fadeout(3)
+            )
             bgm_clip = afx.audio_loop(bgm_clip, duration=video_clip.duration)
             audio_clip = CompositeAudioClip([audio_clip, bgm_clip])
         except Exception as e:
             logger.error(f"failed to add bgm: {str(e)}")
 
     video_clip = video_clip.set_audio(audio_clip)
-    video_clip.write_videofile(output_file,
-                               audio_codec="aac",
-                               temp_audiofile_path=output_dir,
-                               threads=params.n_threads or 2,
-                               logger=None,
-                               fps=30,
-                               )
+    video_clip.write_videofile(
+        output_file,
+        audio_codec="aac",
+        temp_audiofile_path=output_dir,
+        threads=params.n_threads or 2,
+        logger=None,
+        fps=30,
+    )
     video_clip.close()
-    logger.success(f"completed")
+    del video_clip
+    logger.success("completed")
+
+
+def preprocess_video(materials: List[MaterialInfo], clip_duration=4):
+    for material in materials:
+        if not material.url:
+            continue
+
+        ext = utils.parse_extension(material.url)
+        try:
+            clip = VideoFileClip(material.url)
+        except Exception:
+            clip = ImageClip(material.url)
+
+        width = clip.size[0]
+        height = clip.size[1]
+        if width < 480 or height < 480:
+            logger.warning(f"video is too small, width: {width}, height: {height}")
+            continue
+
+        if ext in const.FILE_TYPE_IMAGES:
+            logger.info(f"processing image: {material.url}")
+            # 创建一个图片剪辑，并设置持续时间为3秒钟
+            clip = (
+                ImageClip(material.url)
+                .set_duration(clip_duration)
+                .set_position("center")
+            )
+            # 使用resize方法来添加缩放效果。这里使用了lambda函数来使得缩放效果随时间变化。
+            # 假设我们想要从原始大小逐渐放大到120%的大小。
+            # t代表当前时间，clip.duration为视频总时长，这里是3秒。
+            # 注意：1 表示100%的大小，所以1.2表示120%的大小
+            zoom_clip = clip.resize(
+                lambda t: 1 + (clip_duration * 0.03) * (t / clip.duration)
+            )
+
+            # 如果需要，可以创建一个包含缩放剪辑的复合视频剪辑
+            # （这在您想要在视频中添加其他元素时非常有用）
+            final_clip = CompositeVideoClip([zoom_clip])
+
+            # 输出视频
+            video_file = f"{material.url}.mp4"
+            final_clip.write_videofile(video_file, fps=30, logger=None)
+            final_clip.close()
+            del final_clip
+            material.url = video_file
+            logger.success(f"completed: {video_file}")
+    return materials
 
 
 if __name__ == "__main__":
-    txt_en = "Here's your guide to travel hacks for budget-friendly adventures"
-    txt_zh = "测试长字段这是您的旅行技巧指南帮助您进行预算友好的冒险"
-    font = utils.resource_dir() + "/fonts/STHeitiMedium.ttc"
-    for txt in [txt_en, txt_zh]:
-        t, h = wrap_text(text=txt, max_width=1000, font=font, fontsize=60)
-        print(t)
+    m = MaterialInfo()
+    m.url = "/Users/harry/Downloads/IMG_2915.JPG"
+    m.provider = "local"
+    materials = preprocess_video([m], clip_duration=4)
+    print(materials)
 
-    task_id = "aa563149-a7ea-49c2-b39f-8c32cc225baf"
-    task_dir = utils.task_dir(task_id)
-    video_file = f"{task_dir}/combined-1.mp4"
-    audio_file = f"{task_dir}/audio.mp3"
-    subtitle_file = f"{task_dir}/subtitle.srt"
-    output_file = f"{task_dir}/final.mp4"
-
-    # video_paths = []
-    # for file in os.listdir(utils.storage_dir("test")):
-    #     if file.endswith(".mp4"):
-    #         video_paths.append(os.path.join(utils.storage_dir("test"), file))
+    # txt_en = "Here's your guide to travel hacks for budget-friendly adventures"
+    # txt_zh = "测试长字段这是您的旅行技巧指南帮助您进行预算友好的冒险"
+    # font = utils.resource_dir() + "/fonts/STHeitiMedium.ttc"
+    # for txt in [txt_en, txt_zh]:
+    #     t, h = wrap_text(text=txt, max_width=1000, font=font, fontsize=60)
+    #     print(t)
     #
-    # combine_videos(combined_video_path=video_file,
-    #                audio_file=audio_file,
-    #                video_paths=video_paths,
-    #                video_aspect=VideoAspect.portrait,
-    #                video_concat_mode=VideoConcatMode.random,
-    #                max_clip_duration=5,
-    #                threads=2)
-
-    cfg = VideoParams()
-    cfg.video_aspect = VideoAspect.portrait
-    cfg.font_name = "STHeitiMedium.ttc"
-    cfg.font_size = 60
-    cfg.stroke_color = "#000000"
-    cfg.stroke_width = 1.5
-    cfg.text_fore_color = "#FFFFFF"
-    cfg.text_background_color = "transparent"
-    cfg.bgm_type = "random"
-    cfg.bgm_file = ""
-    cfg.bgm_volume = 1.0
-    cfg.subtitle_enabled = True
-    cfg.subtitle_position = "bottom"
-    cfg.n_threads = 2
-    cfg.paragraph_number = 1
-
-    cfg.voice_volume = 1.0
-
-    generate_video(video_path=video_file,
-                   audio_path=audio_file,
-                   subtitle_path=subtitle_file,
-                   output_file=output_file,
-                   params=cfg
-                   )
+    # task_id = "aa563149-a7ea-49c2-b39f-8c32cc225baf"
+    # task_dir = utils.task_dir(task_id)
+    # video_file = f"{task_dir}/combined-1.mp4"
+    # audio_file = f"{task_dir}/audio.mp3"
+    # subtitle_file = f"{task_dir}/subtitle.srt"
+    # output_file = f"{task_dir}/final.mp4"
+    #
+    # # video_paths = []
+    # # for file in os.listdir(utils.storage_dir("test")):
+    # #     if file.endswith(".mp4"):
+    # #         video_paths.append(os.path.join(utils.storage_dir("test"), file))
+    # #
+    # # combine_videos(combined_video_path=video_file,
+    # #                audio_file=audio_file,
+    # #                video_paths=video_paths,
+    # #                video_aspect=VideoAspect.portrait,
+    # #                video_concat_mode=VideoConcatMode.random,
+    # #                max_clip_duration=5,
+    # #                threads=2)
+    #
+    # cfg = VideoParams()
+    # cfg.video_aspect = VideoAspect.portrait
+    # cfg.font_name = "STHeitiMedium.ttc"
+    # cfg.font_size = 60
+    # cfg.stroke_color = "#000000"
+    # cfg.stroke_width = 1.5
+    # cfg.text_fore_color = "#FFFFFF"
+    # cfg.text_background_color = "transparent"
+    # cfg.bgm_type = "random"
+    # cfg.bgm_file = ""
+    # cfg.bgm_volume = 1.0
+    # cfg.subtitle_enabled = True
+    # cfg.subtitle_position = "bottom"
+    # cfg.n_threads = 2
+    # cfg.paragraph_number = 1
+    #
+    # cfg.voice_volume = 1.0
+    #
+    # generate_video(video_path=video_file,
+    #                audio_path=audio_file,
+    #                subtitle_path=subtitle_file,
+    #                output_file=output_file,
+    #                params=cfg
+    #                )

app/services/voice.py

@@ -15,7 +15,7 @@ from app.utils import utils
 
 def get_all_azure_voices(filter_locals=None) -> list[str]:
     if filter_locals is None:
-        filter_locals = ["zh-CN", "en-US", "zh-HK", "zh-TW"]
+        filter_locals = ["zh-CN", "en-US", "zh-HK", "zh-TW", "vi-VN"]
     voices_str = """
 Name: af-ZA-AdriNeural
 Gender: Female
@@ -988,7 +988,7 @@ Name: zh-CN-XiaoxiaoMultilingualNeural-V2
 Gender: Female
     """.strip()
     voices = []
-    name = ''
+    name = ""
     for line in voices_str.split("\n"):
         line = line.strip()
         if not line:
@@ -1008,7 +1008,7 @@ Gender: Female
                             voices.append(f"{name}-{gender}")
                 else:
                     voices.append(f"{name}-{gender}")
-                name = ''
+                name = ""
     voices.sort()
     return voices
 
@@ -1023,33 +1023,50 @@ def parse_voice_name(name: str):
 
 def is_azure_v2_voice(voice_name: str):
     voice_name = parse_voice_name(voice_name)
-    print(voice_name)
     if voice_name.endswith("-V2"):
         return voice_name.replace("-V2", "").strip()
     return ""
 
 
-def tts(text: str, voice_name: str, voice_file: str) -> [SubMaker, None]:
+def tts(
+    text: str, voice_name: str, voice_rate: float, voice_file: str
+) -> [SubMaker, None]:
     if is_azure_v2_voice(voice_name):
         return azure_tts_v2(text, voice_name, voice_file)
-    return azure_tts_v1(text, voice_name, voice_file)
+    return azure_tts_v1(text, voice_name, voice_rate, voice_file)
 
 
-def azure_tts_v1(text: str, voice_name: str, voice_file: str) -> [SubMaker, None]:
+def convert_rate_to_percent(rate: float) -> str:
+    if rate == 1.0:
+        return "+0%"
+    percent = round((rate - 1.0) * 100)
+    if percent > 0:
+        return f"+{percent}%"
+    else:
+        return f"{percent}%"
+
+
+def azure_tts_v1(
+    text: str, voice_name: str, voice_rate: float, voice_file: str
+) -> [SubMaker, None]:
+    voice_name = parse_voice_name(voice_name)
     text = text.strip()
+    rate_str = convert_rate_to_percent(voice_rate)
     for i in range(3):
         try:
             logger.info(f"start, voice name: {voice_name}, try: {i + 1}")
 
             async def _do() -> SubMaker:
-                communicate = edge_tts.Communicate(text, voice_name)
+                communicate = edge_tts.Communicate(text, voice_name, rate=rate_str)
                 sub_maker = edge_tts.SubMaker()
                 with open(voice_file, "wb") as file:
                     async for chunk in communicate.stream():
                         if chunk["type"] == "audio":
                             file.write(chunk["data"])
                         elif chunk["type"] == "WordBoundary":
-                            sub_maker.create_sub((chunk["offset"], chunk["duration"]), chunk["text"])
+                            sub_maker.create_sub(
+                                (chunk["offset"], chunk["duration"]), chunk["text"]
+                            )
                 return sub_maker
 
             sub_maker = asyncio.run(_do())
@@ -1074,8 +1091,12 @@ def azure_tts_v2(text: str, voice_name: str, voice_file: str) -> [SubMaker, None
     def _format_duration_to_offset(duration) -> int:
         if isinstance(duration, str):
             time_obj = datetime.strptime(duration, "%H:%M:%S.%f")
-            milliseconds = (time_obj.hour * 3600000) + (time_obj.minute * 60000) + (time_obj.second * 1000) + (
-                    time_obj.microsecond // 1000)
+            milliseconds = (
+                (time_obj.hour * 3600000)
+                + (time_obj.minute * 60000)
+                + (time_obj.second * 1000)
+                + (time_obj.microsecond // 1000)
+            )
             return milliseconds * 10000
 
         if isinstance(duration, int):
@@ -1108,20 +1129,29 @@ def azure_tts_v2(text: str, voice_name: str, voice_file: str) -> [SubMaker, None
             # Creates an instance of a speech config with specified subscription key and service region.
             speech_key = config.azure.get("speech_key", "")
             service_region = config.azure.get("speech_region", "")
-            audio_config = speechsdk.audio.AudioOutputConfig(filename=voice_file, use_default_speaker=True)
-            speech_config = speechsdk.SpeechConfig(subscription=speech_key,
-                                                   region=service_region)
+            audio_config = speechsdk.audio.AudioOutputConfig(
+                filename=voice_file, use_default_speaker=True
+            )
+            speech_config = speechsdk.SpeechConfig(
+                subscription=speech_key, region=service_region
+            )
             speech_config.speech_synthesis_voice_name = voice_name
             # speech_config.set_property(property_id=speechsdk.PropertyId.SpeechServiceResponse_RequestSentenceBoundary,
             #                            value='true')
-            speech_config.set_property(property_id=speechsdk.PropertyId.SpeechServiceResponse_RequestWordBoundary,
-                                       value='true')
+            speech_config.set_property(
+                property_id=speechsdk.PropertyId.SpeechServiceResponse_RequestWordBoundary,
+                value="true",
+            )
 
             speech_config.set_speech_synthesis_output_format(
-                speechsdk.SpeechSynthesisOutputFormat.Audio48Khz192KBitRateMonoMp3)
-            speech_synthesizer = speechsdk.SpeechSynthesizer(audio_config=audio_config,
-                                                             speech_config=speech_config)
-            speech_synthesizer.synthesis_word_boundary.connect(speech_synthesizer_word_boundary_cb)
+                speechsdk.SpeechSynthesisOutputFormat.Audio48Khz192KBitRateMonoMp3
+            )
+            speech_synthesizer = speechsdk.SpeechSynthesizer(
+                audio_config=audio_config, speech_config=speech_config
+            )
+            speech_synthesizer.synthesis_word_boundary.connect(
+                speech_synthesizer_word_boundary_cb
+            )
 
             result = speech_synthesizer.speak_text_async(text).get()
             if result.reason == speechsdk.ResultReason.SynthesizingAudioCompleted:
@@ -1129,9 +1159,13 @@ def azure_tts_v2(text: str, voice_name: str, voice_file: str) -> [SubMaker, None
                 return sub_maker
             elif result.reason == speechsdk.ResultReason.Canceled:
                 cancellation_details = result.cancellation_details
-                logger.error(f"azure v2 speech synthesis canceled: {cancellation_details.reason}")
+                logger.error(
+                    f"azure v2 speech synthesis canceled: {cancellation_details.reason}"
+                )
                 if cancellation_details.reason == speechsdk.CancellationReason.Error:
-                    logger.error(f"azure v2 speech synthesis error: {cancellation_details.error_details}")
+                    logger.error(
+                        f"azure v2 speech synthesis error: {cancellation_details.error_details}"
+                    )
             logger.info(f"completed, output file: {voice_file}")
         except Exception as e:
             logger.error(f"failed, error: {str(e)}")
@@ -1168,11 +1202,7 @@ def create_subtitle(sub_maker: submaker.SubMaker, text: str, subtitle_file: str)
         """
         start_t = mktimestamp(start_time).replace(".", ",")
         end_t = mktimestamp(end_time).replace(".", ",")
-        return (
-            f"{idx}\n"
-            f"{start_t} --> {end_t}\n"
-            f"{sub_text}\n"
-        )
+        return f"{idx}\n" f"{start_t} --> {end_t}\n" f"{sub_text}\n"
 
     start_time = -1.0
     sub_items = []
@@ -1229,12 +1259,16 @@ def create_subtitle(sub_maker: submaker.SubMaker, text: str, subtitle_file: str)
             try:
                 sbs = subtitles.file_to_subtitles(subtitle_file, encoding="utf-8")
                 duration = max([tb for ((ta, tb), txt) in sbs])
-                logger.info(f"completed, subtitle file created: {subtitle_file}, duration: {duration}")
+                logger.info(
+                    f"completed, subtitle file created: {subtitle_file}, duration: {duration}"
+                )
             except Exception as e:
                 logger.error(f"failed, error: {str(e)}")
                 os.remove(subtitle_file)
         else:
-            logger.warning(f"failed, sub_items len: {len(sub_items)}, script_lines len: {len(script_lines)}")
+            logger.warning(
+                f"failed, sub_items len: {len(sub_items)}, script_lines len: {len(script_lines)}"
+            )
 
     except Exception as e:
         logger.error(f"failed, error: {str(e)}")
@@ -1258,7 +1292,6 @@ if __name__ == "__main__":
     voices = get_all_azure_voices()
     print(len(voices))
 
-
     async def _do():
         temp_dir = utils.storage_dir("temp")
 
@@ -1307,12 +1340,13 @@ if __name__ == "__main__":
         for voice_name in voice_names:
             voice_file = f"{temp_dir}/tts-{voice_name}.mp3"
             subtitle_file = f"{temp_dir}/tts.mp3.srt"
-            sub_maker = azure_tts_v2(text=text, voice_name=voice_name, voice_file=voice_file)
+            sub_maker = azure_tts_v2(
+                text=text, voice_name=voice_name, voice_file=voice_file
+            )
             create_subtitle(sub_maker=sub_maker, text=text, subtitle_file=subtitle_file)
             audio_duration = get_audio_duration(sub_maker)
             print(f"voice: {voice_name}, audio duration: {audio_duration}s")
 
-
     loop = asyncio.get_event_loop_policy().get_event_loop()
     try:
         loop.run_until_complete(_do())

app/utils/utils.py

@@ -15,12 +15,12 @@ urllib3.disable_warnings()
 
 def get_response(status: int, data: Any = None, message: str = ""):
     obj = {
-        'status': status,
+        "status": status,
     }
     if data:
-        obj['data'] = data
+        obj["data"] = data
     if message:
-        obj['message'] = message
+        obj["message"] = message
     return obj
 
 
@@ -41,7 +41,7 @@ def to_json(obj):
             elif isinstance(o, (list, tuple)):
                 return [serialize(item) for item in o]
             # 如果对象是自定义类型，尝试返回其__dict__属性
-            elif hasattr(o, '__dict__'):
+            elif hasattr(o, "__dict__"):
                 return serialize(o.__dict__)
             # 其他情况返回None（或者可以选择抛出异常）
             else:
@@ -67,10 +67,13 @@ def root_dir():
     return os.path.dirname(os.path.dirname(os.path.dirname(os.path.realpath(__file__))))
 
 
-def storage_dir(sub_dir: str = ""):
+def storage_dir(sub_dir: str = "", create: bool = False):
     d = os.path.join(root_dir(), "storage")
     if sub_dir:
         d = os.path.join(d, sub_dir)
+    if create and not os.path.exists(d):
+        os.makedirs(d)
+
     return d
 
 
@@ -196,7 +199,8 @@ def split_string_by_punctuations(s):
 
 def md5(text):
     import hashlib
-    return hashlib.md5(text.encode('utf-8')).hexdigest()
+
+    return hashlib.md5(text.encode("utf-8")).hexdigest()
 
 
 def get_system_locale():
@@ -219,3 +223,7 @@ def load_locales(i18n_dir):
                 with open(os.path.join(root, file), "r", encoding="utf-8") as f:
                     _locales[lang] = json.loads(f.read())
     return _locales
+
+
+def parse_extension(filename):
+    return os.path.splitext(filename)[1].strip().lower().replace(".", "")

changelog.py

@@ -0,0 +1,17 @@
+from git_changelog.cli import build_and_render
+
+# 运行这段脚本自动生成CHANGELOG.md文件
+
+build_and_render(
+    repository=".",
+    output="CHANGELOG.md",
+    convention="angular",
+    provider="github",
+    template="keepachangelog",
+    parse_trailers=True,
+    parse_refs=False,
+    sections=["build", "deps", "feat", "fix", "refactor"],
+    versioning="pep440",
+    bump="1.1.2",  # 指定bump版本
+    in_place=True,
+)

config.example.toml

@@ -1,4 +1,6 @@
 [app]
+
+    video_source = "pexels"  # "pexels" or "pixabay"
     # Pexels API Key
     # Register at https://www.pexels.com/api/ to get your API key.
     # You can use multiple keys to avoid rate limits.
@@ -6,6 +8,13 @@
     # 特别注意格式，Key 用英文双引号括起来，多个Key用逗号隔开
     pexels_api_keys = []
 
+    # Pixabay API Key
+    # Register at https://pixabay.com/api/docs/ to get your API key.
+    # You can use multiple keys to avoid rate limits.
+    # For example: pixabay_api_keys = ["123adsf4567adf89","abd1321cd13efgfdfhi"]
+    # 特别注意格式，Key 用英文双引号括起来，多个Key用逗号隔开
+    pixabay_api_keys = []
+
     # 如果你没有 OPENAI API Key，可以使用 g4f 代替，或者使用国内的 Moonshot API
     # If you don't have an OPENAI API Key, you can use g4f instead
 
@@ -31,7 +40,7 @@
     # No need to set it unless you want to use your own proxy
     openai_base_url = ""
     # Check your available models at https://platform.openai.com/account/limits
-    openai_model_name = "gpt-4-turbo-preview"
+    openai_model_name = "gpt-4-turbo"
 
     ########## Moonshot API Key
     # Visit https://platform.moonshot.cn/console/api-keys to get your API key.
@@ -48,7 +57,7 @@
     ########## G4F
     # Visit https://github.com/xtekky/gpt4free to get more details
     # Supported model list: https://github.com/xtekky/gpt4free/blob/main/g4f/models.py
-    g4f_model_name = "gpt-3.5-turbo-16k-0613"
+    g4f_model_name = "gpt-3.5-turbo"
 
     ########## Azure API Key
     # Visit https://learn.microsoft.com/zh-cn/azure/ai-services/openai/ to get more details
@@ -68,9 +77,15 @@
     # https://tongyi.aliyun.com/qianwen/
     # https://help.aliyun.com/zh/dashscope/developer-reference/model-introduction
     qwen_api_key = ""
-    qwen_model_name = "qwen-max" 
+    qwen_model_name = "qwen-max"
 
 
+    ########## DeepSeek API Key
+    # Visit https://platform.deepseek.com/api_keys to get your API key
+    deepseek_api_key = ""
+    deepseek_base_url = "https://api.deepseek.com"
+    deepseek_model_name = "deepseek-chat"
+
     # Subtitle Provider, "edge" or "whisper"
     # If empty, the subtitle will not be generated
     subtitle_provider = "edge"
@@ -134,6 +149,15 @@
     redis_host = "localhost"
     redis_port = 6379
     redis_db = 0
+    redis_password = ""
+
+    # 文生视频时的最大并发任务数
+    max_concurrent_tasks = 5
+
+    # webui界面是否显示配置项
+    # webui hide baisc config panel
+    hide_config = false
+
 
 [whisper]
     # Only effective when subtitle_provider is "whisper"
@@ -153,15 +177,15 @@
     device="CPU"
     compute_type="int8"
 
-[pexels]
-    video_concat_mode="sequential" # "random" or "sequential"
-    [pexels.proxies]
-        ### Use a proxy to access the Pexels API
-        ### Format: "http://<username>:<password>@<proxy>:<port>"
-        ### Example: "http://user:pass@proxy:1234"
-        ### Doc: https://requests.readthedocs.io/en/latest/user/advanced/#proxies
-        # http = "http://10.10.1.10:3128"
-        # https = "http://10.10.1.10:1080"
+
+[proxy]
+    ### Use a proxy to access the Pexels API
+    ### Format: "http://<username>:<password>@<proxy>:<port>"
+    ### Example: "http://user:pass@proxy:1234"
+    ### Doc: https://requests.readthedocs.io/en/latest/user/advanced/#proxies
+
+    # http = "http://10.10.1.10:3128"
+    # https = "http://10.10.1.10:1080"
 
 [azure]
     # Azure Speech API Key

main.py

@@ -1,8 +1,16 @@
 import uvicorn
 from loguru import logger
+
 from app.config import config
 
-if __name__ == '__main__':
-    logger.info("start server, docs: http://127.0.0.1:" + str(config.listen_port) + "/docs")
-    uvicorn.run(app="app.asgi:app", host=config.listen_host, port=config.listen_port, reload=config.reload_debug,
-                log_level="warning")
+if __name__ == "__main__":
+    logger.info(
+        "start server, docs: http://127.0.0.1:" + str(config.listen_port) + "/docs"
+    )
+    uvicorn.run(
+        app="app.asgi:app",
+        host=config.listen_host,
+        port=config.listen_port,
+        reload=config.reload_debug,
+        log_level="warning",
+    )

requirements.txt

@@ -6,20 +6,21 @@ edge_tts~=6.1.10
 uvicorn~=0.27.1
 fastapi~=0.110.0
 tomli~=2.0.1
-streamlit~=1.32.0
+streamlit~=1.33.0
 loguru~=0.7.2
 aiohttp~=3.9.3
 urllib3~=2.2.1
-pillow~=9.5.0
+pillow~=10.3.0
 pydantic~=2.6.3
-g4f~=0.2.5.4
+g4f~=0.3.0.4
 dashscope~=1.15.0
 google.generativeai~=0.4.1
 python-multipart~=0.0.9
 redis==5.0.3
 # if you use pillow~=10.3.0, you will get "PIL.Image' has no attribute 'ANTIALIAS'" error when resize video
 # please install opencv-python to fix "PIL.Image' has no attribute 'ANTIALIAS'" error
-opencv-python
+opencv-python~=4.9.0.80
 # for azure speech
 # https://techcommunity.microsoft.com/t5/ai-azure-ai-services-blog/9-more-realistic-ai-voices-for-conversations-now-generally/ba-p/4099471
-azure-cognitiveservices-speech~=1.37.0
+azure-cognitiveservices-speech~=1.37.0
+git-changelog~=2.5.2

sites/docs/.vuepress/config.ts

@@ -0,0 +1,208 @@
+import { viteBundler } from "@vuepress/bundler-vite";
+import { defaultTheme } from "@vuepress/theme-default";
+import { defineUserConfig } from "vuepress";
+
+const base = "MoneyPrinterTurbo";
+const isProd = process.env.NODE_ENV === "production";
+
+export default defineUserConfig({
+  lang: "zh-CN",
+  base: `/${base}/`,
+  bundler: viteBundler(),
+  theme: defaultTheme({
+    repo: "harry0703/MoneyPrinterTurbo",
+    docsDir: "sites/docs",
+    colorModeSwitch: true,
+    locales: {
+      "/": {
+        // navbar
+        navbar: [
+          { text: "Guide", link: "/guide/" },
+          // { text: "Components", link: "/components/" },
+        ],
+        selectLanguageText: "Languages",
+        selectLanguageName: "English",
+        selectLanguageAriaLabel: "Select language",
+        // sidebar
+        sidebar: {
+          "/guide/": [
+            {
+              text: "Guide",
+              children: [
+                { text: "Get Started", link: "/guide/README.md" }, 
+                { text: "Video Demonstration", link: "/guide/video-demonstration.md" },
+                { text: "Features", link: "/guide/features.md" },
+                { text: "Speech Synthesis", link: "/guide/speech-synthesis.md" },
+                { text: "Subtitle Generation", link: "/guide/subtitle-generation.md" },
+                { text: "Background Music", link: "/guide/background-music.md" },
+                { text: "Subtitle Font", link: "/guide/subtitle-font.md" },
+              ],
+            },
+            {
+              text: "Others",
+              children: [
+                { text: "FAQ", link: "/guide/faq.md" },
+                { text: "Feedback", link: "/guide/feedback.md" },
+                { text: "Reference Project", link: "/guide/reference-project.md" },
+              ],
+            },
+          ],
+          // "/components/": getComponentsSidebar("Components", "Advanced"),
+        },
+        // page meta
+        editLinkText: "Edit this page on GitHub",
+      },
+      "/zh/": {
+        // navbar
+        navbar: [
+          { text: "指南", link: "/zh/guide/" },
+          // { text: "组件", link: "/zh/components/" },
+        ],
+        selectLanguageText: "选择语言",
+        selectLanguageName: "简体中文",
+        selectLanguageAriaLabel: "选择语言",
+        // sidebar
+        sidebar: {
+          "/zh/guide/": [
+            {
+              text: "指南",
+              children: [
+                { text: "快速开始", link: "/zh/guide/README.md" }, 
+                { text: "配置要求", link: "/zh/guide/configuration-requirements.md" },
+                { text: "视频演示", link: "/zh/guide/video-demonstration.md" },
+                { text: "功能", link: "/zh/guide/features.md" },
+                { text: "语音合成", link: "/zh/guide/speech-synthesis.md" },
+                { text: "字幕生成", link: "/zh/guide/subtitle-generation.md" },
+                { text: "背景音乐", link: "/zh/guide/background-music.md" },
+                { text: "字幕字体", link: "/zh/guide/subtitle-font.md" },
+              ],
+            },
+            {
+              text: "其他",
+              children: [
+                { text: "常见问题", link: "/zh/guide/faq.md" },
+                { text: "反馈建议", link: "/zh/guide/feedback.md" },
+                { text: "参考项目", link: "/zh/guide/reference-project.md" },
+                { text: "特别感谢", link: "/zh/guide/special-thanks.md" }, 
+                { text: "感谢赞助", link: "/zh/guide/thanks-for-sponsoring" },
+              ],
+            },
+          ],
+          // "/zh/others/": getComponentsSidebar("组件", "高级"),
+        },
+        // page meta
+        editLinkText: "在 GitHub 上编辑此页",
+        lastUpdatedText: "上次更新",
+        contributorsText: "贡献者",
+        // custom containers
+        tip: "提示",
+        warning: "注意",
+        danger: "警告",
+        // 404 page
+        notFound: [
+          "这里什么都没有",
+          "我们怎么到这来了？",
+          "这是一个 404 页面",
+          "看起来我们进入了错误的链接",
+        ],
+        backToHome: "返回首页",
+      },
+    },
+    themePlugins: {
+      // only enable git plugin in production mode
+      git: isProd,
+    },
+  }),
+  locales: {
+    "/": {
+      lang: "en-US",
+      title: "MoneyPrinterTurbo",
+      description: "Generate short videos with one click using AI LLM.",
+    },
+    "/zh/": {
+      lang: "zh-CN",
+      title: "MoneyPrinterTurbo",
+      description: "利用AI大模型，一键生成高清短视频。",
+    },
+  },
+  head: [
+    [
+      "link",
+      {
+        rel: "icon",
+        type: "image/png",
+        sizes: "16x16",
+        href: `/${base}/icons/favicon-16x16.png`,
+      },
+    ],
+    [
+      "link",
+      {
+        rel: "icon",
+        type: "image/png",
+        sizes: "32x32",
+        href: `/${base}/icons/favicon-32x32.png`,
+      },
+    ],
+    ["meta", { name: "application-name", content: "MoneyPrinterTurbo" }],
+    [
+      "meta",
+      { name: "apple-mobile-web-app-title", content: "MoneyPrinterTurbo" },
+    ],
+    ["meta", { name: "apple-mobile-web-app-capable", content: "yes" }],
+    [
+      "meta",
+      { name: "apple-mobile-web-app-status-bar-style", content: "black" },
+    ],
+    [
+      "link",
+      {
+        rel: "apple-touch-icon",
+        href: `/${base}/icons/apple-touch-icon-152x152.png`,
+      },
+    ],
+    [
+      "link",
+      {
+        rel: "mask-icon",
+        href: "/${base}/icons/safari-pinned-tab.svg",
+        color: "#3eaf7c",
+      },
+    ],
+    [
+      "meta",
+      {
+        name: "msapplication-TileImage",
+        content: "/${base}/icons/msapplication-icon-144x144.png",
+      },
+    ],
+    ["meta", { name: "msapplication-TileColor", content: "#000000" }],
+    ["meta", { name: "theme-color", content: "#3eaf7c" }],
+  ],
+});
+
+function getGuideSidebar(groupA: string, groupB: string) {
+  return [
+    {
+      text: groupA,
+      children: ["README.md", { text: "特别感谢", link: "/zh/guide/special-thanks.md" }, "2.md"],
+    },
+    {
+      text: groupB,
+      children: ["custom-validator.md", "1.md", "2.md", "3.md"],
+    },
+  ];
+}
+
+function getComponentsSidebar(groupA: string, groupB: string) {
+  return [
+    {
+      text: groupA,
+      children: ["README.md", "1.md", "2.md"],
+    },
+    {
+      text: groupB,
+      children: ["custom-components.md"],
+    },
+  ];
+}

sites/docs/.vuepress/public/icons/safari-pinned-tab.svg

@@ -0,0 +1,149 @@
+<?xml version="1.0" standalone="no"?>
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 20010904//EN"
+ "http://www.w3.org/TR/2001/REC-SVG-20010904/DTD/svg10.dtd">
+<svg version="1.0" xmlns="http://www.w3.org/2000/svg"
+ width="16.000000pt" height="16.000000pt" viewBox="0 0 16.000000 16.000000"
+ preserveAspectRatio="xMidYMid meet">
+<metadata>
+Created by potrace 1.11, written by Peter Selinger 2001-2013
+</metadata>
+<g transform="translate(0.000000,16.000000) scale(0.000320,-0.000320)"
+fill="#000000" stroke="none">
+<path d="M18 46618 c45 -75 122 -207 122 -211 0 -2 25 -45 55 -95 30 -50 55
+-96 55 -102 0 -5 5 -10 10 -10 6 0 10 -4 10 -9 0 -5 73 -135 161 -288 89 -153
+173 -298 187 -323 14 -25 32 -57 41 -72 88 -149 187 -324 189 -335 2 -7 8 -13
+13 -13 5 0 9 -4 9 -10 0 -5 46 -89 103 -187 175 -302 490 -846 507 -876 8 -16
+20 -36 25 -45 28 -46 290 -498 339 -585 13 -23 74 -129 136 -236 61 -107 123
+-215 137 -240 14 -25 29 -50 33 -56 5 -5 23 -37 40 -70 18 -33 38 -67 44 -75
+11 -16 21 -33 63 -109 14 -25 29 -50 33 -56 4 -5 21 -35 38 -65 55 -100 261
+-455 269 -465 4 -5 14 -21 20 -35 15 -29 41 -75 103 -180 24 -41 52 -88 60
+-105 9 -16 57 -100 107 -185 112 -193 362 -626 380 -660 8 -14 23 -38 33 -55
+11 -16 23 -37 27 -45 4 -8 26 -46 48 -85 23 -38 53 -90 67 -115 46 -81 64
+-113 178 -310 62 -107 121 -210 132 -227 37 -67 56 -99 85 -148 16 -27 32 -57
+36 -65 4 -8 15 -27 25 -42 9 -15 53 -89 96 -165 44 -76 177 -307 296 -513 120
+-206 268 -463 330 -570 131 -227 117 -203 200 -348 36 -62 73 -125 82 -140 10
+-15 21 -34 25 -42 4 -8 20 -37 36 -65 17 -27 38 -65 48 -82 49 -85 64 -111 87
+-153 13 -25 28 -49 32 -55 4 -5 78 -134 165 -285 87 -151 166 -288 176 -305
+10 -16 26 -43 35 -59 9 -17 125 -217 257 -445 132 -229 253 -441 270 -471 17
+-30 45 -79 64 -108 18 -29 33 -54 33 -57 0 -2 20 -37 44 -77 24 -40 123 -212
+221 -383 97 -170 190 -330 205 -355 16 -25 39 -65 53 -90 13 -25 81 -144 152
+-265 70 -121 137 -238 150 -260 12 -22 37 -65 55 -95 18 -30 43 -73 55 -95 12
+-22 48 -85 80 -140 77 -132 163 -280 190 -330 13 -22 71 -123 130 -225 59
+-102 116 -199 126 -217 10 -17 29 -50 43 -72 15 -22 26 -43 26 -45 0 -2 27
+-50 60 -106 33 -56 60 -103 60 -105 0 -2 55 -98 90 -155 8 -14 182 -316 239
+-414 13 -22 45 -79 72 -124 27 -46 49 -86 49 -89 0 -2 14 -24 30 -48 16 -24
+30 -46 30 -49 0 -5 74 -135 100 -176 5 -8 24 -42 43 -75 50 -88 58 -101 262
+-455 104 -179 199 -345 213 -370 14 -25 28 -49 32 -55 4 -5 17 -26 28 -45 10
+-19 62 -109 114 -200 114 -197 133 -230 170 -295 16 -27 33 -57 38 -65 17 -28
+96 -165 103 -180 4 -8 16 -28 26 -45 10 -16 77 -131 148 -255 72 -124 181
+-313 243 -420 62 -107 121 -209 131 -227 35 -62 323 -560 392 -678 38 -66 83
+-145 100 -175 16 -30 33 -59 37 -65 4 -5 17 -27 29 -47 34 -61 56 -100 90
+-156 17 -29 31 -55 31 -57 0 -2 17 -32 39 -67 21 -35 134 -229 251 -433 117
+-203 235 -407 261 -451 27 -45 49 -85 49 -88 0 -4 8 -19 19 -34 15 -21 200
+-341 309 -533 10 -19 33 -58 51 -87 17 -29 31 -54 31 -56 0 -2 25 -44 55 -94
+30 -50 55 -95 55 -98 0 -4 6 -15 14 -23 7 -9 27 -41 43 -71 17 -30 170 -297
+342 -594 171 -296 311 -542 311 -547 0 -5 5 -9 10 -9 6 0 10 -4 10 -10 0 -5
+22 -47 49 -92 27 -46 58 -99 68 -118 24 -43 81 -140 93 -160 5 -8 66 -114 135
+-235 69 -121 130 -227 135 -235 12 -21 259 -447 283 -490 10 -19 28 -47 38
+-62 11 -14 19 -29 19 -32 0 -3 37 -69 83 -148 99 -170 305 -526 337 -583 13
+-22 31 -53 41 -70 11 -16 22 -37 26 -45 7 -14 82 -146 103 -180 14 -24 181
+-311 205 -355 13 -22 46 -80 75 -130 29 -49 64 -110 78 -135 14 -25 51 -88 82
+-140 31 -52 59 -102 63 -110 4 -8 18 -33 31 -55 205 -353 284 -489 309 -535
+17 -30 45 -78 62 -106 18 -28 36 -60 39 -72 4 -12 12 -22 17 -22 5 0 9 -4 9
+-10 0 -5 109 -197 241 -427 133 -230 250 -431 259 -448 51 -90 222 -385 280
+-485 37 -63 78 -135 92 -160 14 -25 67 -117 118 -205 51 -88 101 -175 111
+-193 34 -58 55 -95 149 -257 51 -88 101 -173 110 -190 9 -16 76 -131 147 -255
+72 -124 140 -241 151 -260 61 -108 281 -489 355 -615 38 -66 77 -133 87 -150
+35 -63 91 -161 100 -175 14 -23 99 -169 128 -220 54 -97 135 -235 142 -245 4
+-5 20 -32 35 -60 26 -48 238 -416 276 -480 10 -16 26 -46 37 -65 30 -53 382
+-661 403 -695 10 -16 22 -37 26 -45 4 -8 26 -48 50 -88 24 -41 43 -75 43 -77
+0 -2 22 -40 50 -85 27 -45 50 -84 50 -86 0 -3 38 -69 83 -147 84 -142 302
+-520 340 -587 10 -19 34 -60 52 -90 18 -30 44 -75 57 -100 14 -25 45 -79 70
+-120 25 -41 56 -96 70 -121 14 -25 77 -133 138 -240 62 -107 122 -210 132
+-229 25 -43 310 -535 337 -581 11 -19 26 -45 34 -59 17 -32 238 -414 266 -460
+11 -19 24 -41 28 -49 3 -7 75 -133 160 -278 84 -146 153 -269 153 -274 0 -5 5
+-9 10 -9 6 0 10 -4 10 -10 0 -5 82 -150 181 -322 182 -314 201 -346 240 -415
+12 -21 80 -139 152 -263 71 -124 141 -245 155 -270 14 -25 28 -49 32 -55 6 -8
+145 -248 220 -380 37 -66 209 -362 229 -395 11 -19 24 -42 28 -49 4 -8 67
+-118 140 -243 73 -125 133 -230 133 -233 0 -2 15 -28 33 -57 19 -29 47 -78 64
+-108 17 -30 53 -93 79 -139 53 -90 82 -141 157 -272 82 -142 115 -199 381
+-659 142 -245 268 -463 281 -485 12 -22 71 -125 132 -230 60 -104 172 -298
+248 -430 76 -132 146 -253 156 -270 11 -16 22 -36 26 -44 3 -8 30 -54 60 -103
+29 -49 53 -91 53 -93 0 -3 18 -34 40 -70 22 -36 40 -67 40 -69 0 -2 37 -66 81
+-142 45 -77 98 -168 119 -204 20 -36 47 -81 58 -100 12 -19 27 -47 33 -62 6
+-16 15 -28 20 -28 5 0 9 -4 9 -9 0 -6 63 -118 140 -251 77 -133 140 -243 140
+-245 0 -2 18 -33 41 -70 22 -37 49 -83 60 -101 10 -19 29 -51 40 -71 25 -45
+109 -189 126 -218 7 -11 17 -29 22 -40 6 -11 22 -38 35 -60 14 -22 37 -62 52
+-90 14 -27 35 -62 45 -77 11 -14 19 -29 19 -32 0 -3 18 -35 40 -71 22 -36 40
+-67 40 -69 0 -2 19 -35 42 -72 23 -38 55 -94 72 -124 26 -47 139 -244 171
+-298 6 -9 21 -36 34 -60 28 -48 37 -51 51 -19 6 12 19 36 29 52 10 17 27 46
+38 65 11 19 104 181 208 360 103 179 199 345 213 370 14 25 42 74 64 109 21
+34 38 65 38 67 0 2 18 33 40 69 22 36 40 67 40 69 0 3 177 310 199 346 16 26
+136 234 140 244 2 5 25 44 52 88 27 44 49 81 49 84 0 2 18 34 40 70 22 36 40
+67 40 69 0 2 20 36 43 77 35 58 169 289 297 513 9 17 50 86 90 155 40 69 86
+150 103 180 16 30 35 62 41 70 6 8 16 24 22 35 35 64 72 129 167 293 59 100
+116 199 127 220 11 20 30 53 41 72 43 72 1070 1850 1121 1940 14 25 65 113
+113 195 48 83 96 166 107 185 10 19 28 50 38 68 11 18 73 124 137 235 64 111
+175 303 246 427 71 124 173 299 225 390 52 91 116 202 143 248 27 45 49 85 49
+89 0 4 6 14 14 22 7 9 28 43 46 76 26 47 251 436 378 655 11 19 29 51 40 70
+11 19 101 176 201 348 99 172 181 317 181 323 0 5 5 9 10 9 6 0 10 5 10 11 0
+6 8 23 18 37 11 15 32 52 49 82 16 30 130 228 253 440 122 212 234 405 248
+430 13 25 39 70 57 100 39 65 69 117 130 225 25 44 50 87 55 95 12 19 78 134
+220 380 61 107 129 224 150 260 161 277 222 382 246 425 15 28 47 83 71 123
+24 41 43 78 43 83 0 5 4 9 8 9 4 0 13 12 19 28 7 15 23 45 36 67 66 110 277
+478 277 483 0 3 6 13 14 21 7 9 27 41 43 71 17 30 45 80 63 110 34 57 375 649
+394 685 6 11 16 27 22 35 6 8 26 42 44 75 18 33 41 74 51 90 10 17 24 41 32
+55 54 97 72 128 88 152 11 14 19 28 19 30 0 3 79 141 175 308 96 167 175 305
+175 308 0 3 6 13 14 21 7 9 26 39 41 66 33 60 276 483 338 587 24 40 46 80 50
+88 4 8 13 24 20 35 14 23 95 163 125 215 11 19 52 91 92 160 40 69 80 139 90
+155 9 17 103 179 207 360 105 182 200 346 211 365 103 181 463 802 489 845 7
+11 15 27 19 35 4 8 29 51 55 95 64 110 828 1433 848 1470 9 17 24 41 33 55 9
+14 29 48 45 77 15 28 52 93 82 145 30 51 62 107 71 123 17 30 231 398 400 690
+51 88 103 179 115 202 12 23 26 48 32 55 6 7 24 38 40 68 17 30 61 107 98 170
+37 63 84 144 103 180 19 36 41 72 48 81 8 8 14 18 14 21 0 4 27 51 59 106 32
+55 72 124 89 154 16 29 71 125 122 213 51 88 104 180 118 205 13 25 28 50 32
+55 4 6 17 26 28 45 11 19 45 80 77 135 31 55 66 116 77 135 11 19 88 152 171
+295 401 694 620 1072 650 1125 11 19 87 152 170 295 83 143 158 273 166 288 9
+16 21 36 26 45 6 9 31 52 55 96 25 43 54 94 66 115 11 20 95 164 186 321 91
+157 173 299 182 315 9 17 26 46 37 65 12 19 66 114 121 210 56 96 108 186 117
+200 8 14 24 40 34 59 24 45 383 664 412 713 5 9 17 29 26 45 15 28 120 210
+241 419 36 61 68 117 72 125 4 8 12 23 19 34 35 57 245 420 262 453 11 20 35
+61 53 90 17 29 32 54 32 56 0 3 28 51 62 108 33 57 70 119 80 138 10 19 23 42
+28 50 5 8 32 53 59 100 27 47 149 258 271 470 122 212 234 405 248 430 30 53
+62 108 80 135 6 11 15 27 19 35 4 8 85 150 181 315 96 165 187 323 202 350 31
+56 116 202 130 225 5 8 25 42 43 75 19 33 92 159 162 280 149 257 157 271 202
+350 19 33 38 67 43 75 9 14 228 392 275 475 12 22 55 96 95 165 40 69 80 139
+90 155 24 42 202 350 221 383 9 15 27 47 41 72 14 25 75 131 136 236 61 106
+121 210 134 232 99 172 271 470 279 482 5 8 23 40 40 70 18 30 81 141 142 245
+60 105 121 210 135 235 14 25 71 124 127 220 56 96 143 247 194 335 51 88 96
+167 102 175 14 24 180 311 204 355 23 43 340 590 356 615 5 8 50 87 101 175
+171 301 517 898 582 1008 25 43 46 81 46 83 0 2 12 23 27 47 14 23 40 67 56
+97 16 30 35 62 42 70 7 8 15 22 18 30 4 8 20 38 37 65 16 28 33 57 37 65 6 12
+111 196 143 250 5 8 55 95 112 193 57 98 113 195 126 215 12 20 27 46 32 57 6
+11 14 27 20 35 5 8 76 130 156 270 80 140 165 287 187 325 23 39 52 90 66 115
+13 25 30 52 37 61 8 8 14 18 14 21 0 4 41 77 92 165 50 87 175 302 276 478
+101 176 208 360 236 408 28 49 67 117 86 152 19 35 41 70 48 77 6 6 12 15 12
+19 0 7 124 224 167 291 12 21 23 40 23 42 0 2 21 40 46 83 26 43 55 92 64 109
+54 95 327 568 354 614 19 30 45 75 59 100 71 128 82 145 89 148 4 2 8 8 8 13
+0 5 42 82 94 172 311 538 496 858 518 897 14 25 40 70 58 100 18 30 42 71 53
+90 10 19 79 139 152 265 73 127 142 246 153 265 10 19 43 76 72 125 29 50 63
+108 75 130 65 116 80 140 87 143 4 2 8 8 8 12 0 8 114 212 140 250 6 8 14 24
+20 35 5 11 54 97 108 190 l100 170 -9611 3 c-5286 1 -9614 -1 -9618 -5 -5 -6
+-419 -719 -619 -1068 -89 -155 -267 -463 -323 -560 -38 -66 -81 -140 -95 -165
+-31 -56 -263 -457 -526 -910 -110 -190 -224 -388 -254 -440 -29 -52 -61 -109
+-71 -125 -23 -39 -243 -420 -268 -465 -11 -19 -204 -352 -428 -740 -224 -388
+-477 -826 -563 -975 -85 -148 -185 -322 -222 -385 -37 -63 -120 -207 -185
+-320 -65 -113 -177 -306 -248 -430 -72 -124 -172 -297 -222 -385 -51 -88 -142
+-245 -202 -350 -131 -226 -247 -427 -408 -705 -65 -113 -249 -432 -410 -710
+-160 -278 -388 -673 -506 -877 -118 -205 -216 -373 -219 -373 -3 0 -52 82
+-109 183 -58 100 -144 250 -192 332 -95 164 -402 696 -647 1120 -85 149 -228
+396 -317 550 -212 365 -982 1700 -1008 1745 -10 19 -43 76 -72 125 -29 50 -64
+110 -77 135 -14 25 -63 110 -110 190 -47 80 -96 165 -110 190 -14 25 -99 171
+-188 325 -89 154 -174 300 -188 325 -13 25 -64 113 -112 195 -48 83 -140 242
+-205 355 -65 113 -183 317 -263 454 -79 137 -152 264 -163 282 -50 89 -335
+583 -354 614 -12 19 -34 58 -50 85 -15 28 -129 226 -253 440 -124 215 -235
+408 -247 430 -12 22 -69 121 -127 220 -58 99 -226 389 -373 645 -148 256 -324
+561 -392 678 -67 117 -134 232 -147 255 -13 23 -33 59 -46 80 l-22 37 -9615 0
+-9615 0 20 -32z"/>
+</g>
+</svg>

sites/docs/.vuepress/public/manifest.webmanifest

@@ -0,0 +1,21 @@
+{
+  "name": "VuePress",
+  "short_name": "VuePress",
+  "description": "Vue-powered Static Site Generator",
+  "start_url": "/index.html",
+  "display": "standalone",
+  "background_color": "#fff",
+  "theme_color": "#3eaf7c",
+  "icons": [
+    {
+      "src": "/icons/android-chrome-192x192.png",
+      "sizes": "192x192",
+      "type": "image/png"
+    },
+    {
+      "src": "/icons/android-chrome-512x512.png",
+      "sizes": "512x512",
+      "type": "image/png"
+    }
+  ]
+}

sites/docs/.vuepress/public/voice-list.txt

@@ -0,0 +1,941 @@
+Name: af-ZA-AdriNeural
+Gender: Female
+
+Name: af-ZA-WillemNeural
+Gender: Male
+
+Name: am-ET-AmehaNeural
+Gender: Male
+
+Name: am-ET-MekdesNeural
+Gender: Female
+
+Name: ar-AE-FatimaNeural
+Gender: Female
+
+Name: ar-AE-HamdanNeural
+Gender: Male
+
+Name: ar-BH-AliNeural
+Gender: Male
+
+Name: ar-BH-LailaNeural
+Gender: Female
+
+Name: ar-DZ-AminaNeural
+Gender: Female
+
+Name: ar-DZ-IsmaelNeural
+Gender: Male
+
+Name: ar-EG-SalmaNeural
+Gender: Female
+
+Name: ar-EG-ShakirNeural
+Gender: Male
+
+Name: ar-IQ-BasselNeural
+Gender: Male
+
+Name: ar-IQ-RanaNeural
+Gender: Female
+
+Name: ar-JO-SanaNeural
+Gender: Female
+
+Name: ar-JO-TaimNeural
+Gender: Male
+
+Name: ar-KW-FahedNeural
+Gender: Male
+
+Name: ar-KW-NouraNeural
+Gender: Female
+
+Name: ar-LB-LaylaNeural
+Gender: Female
+
+Name: ar-LB-RamiNeural
+Gender: Male
+
+Name: ar-LY-ImanNeural
+Gender: Female
+
+Name: ar-LY-OmarNeural
+Gender: Male
+
+Name: ar-MA-JamalNeural
+Gender: Male
+
+Name: ar-MA-MounaNeural
+Gender: Female
+
+Name: ar-OM-AbdullahNeural
+Gender: Male
+
+Name: ar-OM-AyshaNeural
+Gender: Female
+
+Name: ar-QA-AmalNeural
+Gender: Female
+
+Name: ar-QA-MoazNeural
+Gender: Male
+
+Name: ar-SA-HamedNeural
+Gender: Male
+
+Name: ar-SA-ZariyahNeural
+Gender: Female
+
+Name: ar-SY-AmanyNeural
+Gender: Female
+
+Name: ar-SY-LaithNeural
+Gender: Male
+
+Name: ar-TN-HediNeural
+Gender: Male
+
+Name: ar-TN-ReemNeural
+Gender: Female
+
+Name: ar-YE-MaryamNeural
+Gender: Female
+
+Name: ar-YE-SalehNeural
+Gender: Male
+
+Name: az-AZ-BabekNeural
+Gender: Male
+
+Name: az-AZ-BanuNeural
+Gender: Female
+
+Name: bg-BG-BorislavNeural
+Gender: Male
+
+Name: bg-BG-KalinaNeural
+Gender: Female
+
+Name: bn-BD-NabanitaNeural
+Gender: Female
+
+Name: bn-BD-PradeepNeural
+Gender: Male
+
+Name: bn-IN-BashkarNeural
+Gender: Male
+
+Name: bn-IN-TanishaaNeural
+Gender: Female
+
+Name: bs-BA-GoranNeural
+Gender: Male
+
+Name: bs-BA-VesnaNeural
+Gender: Female
+
+Name: ca-ES-EnricNeural
+Gender: Male
+
+Name: ca-ES-JoanaNeural
+Gender: Female
+
+Name: cs-CZ-AntoninNeural
+Gender: Male
+
+Name: cs-CZ-VlastaNeural
+Gender: Female
+
+Name: cy-GB-AledNeural
+Gender: Male
+
+Name: cy-GB-NiaNeural
+Gender: Female
+
+Name: da-DK-ChristelNeural
+Gender: Female
+
+Name: da-DK-JeppeNeural
+Gender: Male
+
+Name: de-AT-IngridNeural
+Gender: Female
+
+Name: de-AT-JonasNeural
+Gender: Male
+
+Name: de-CH-JanNeural
+Gender: Male
+
+Name: de-CH-LeniNeural
+Gender: Female
+
+Name: de-DE-AmalaNeural
+Gender: Female
+
+Name: de-DE-ConradNeural
+Gender: Male
+
+Name: de-DE-FlorianMultilingualNeural
+Gender: Male
+
+Name: de-DE-KatjaNeural
+Gender: Female
+
+Name: de-DE-KillianNeural
+Gender: Male
+
+Name: de-DE-SeraphinaMultilingualNeural
+Gender: Female
+
+Name: el-GR-AthinaNeural
+Gender: Female
+
+Name: el-GR-NestorasNeural
+Gender: Male
+
+Name: en-AU-NatashaNeural
+Gender: Female
+
+Name: en-AU-WilliamNeural
+Gender: Male
+
+Name: en-CA-ClaraNeural
+Gender: Female
+
+Name: en-CA-LiamNeural
+Gender: Male
+
+Name: en-GB-LibbyNeural
+Gender: Female
+
+Name: en-GB-MaisieNeural
+Gender: Female
+
+Name: en-GB-RyanNeural
+Gender: Male
+
+Name: en-GB-SoniaNeural
+Gender: Female
+
+Name: en-GB-ThomasNeural
+Gender: Male
+
+Name: en-HK-SamNeural
+Gender: Male
+
+Name: en-HK-YanNeural
+Gender: Female
+
+Name: en-IE-ConnorNeural
+Gender: Male
+
+Name: en-IE-EmilyNeural
+Gender: Female
+
+Name: en-IN-NeerjaExpressiveNeural
+Gender: Female
+
+Name: en-IN-NeerjaNeural
+Gender: Female
+
+Name: en-IN-PrabhatNeural
+Gender: Male
+
+Name: en-KE-AsiliaNeural
+Gender: Female
+
+Name: en-KE-ChilembaNeural
+Gender: Male
+
+Name: en-NG-AbeoNeural
+Gender: Male
+
+Name: en-NG-EzinneNeural
+Gender: Female
+
+Name: en-NZ-MitchellNeural
+Gender: Male
+
+Name: en-NZ-MollyNeural
+Gender: Female
+
+Name: en-PH-JamesNeural
+Gender: Male
+
+Name: en-PH-RosaNeural
+Gender: Female
+
+Name: en-SG-LunaNeural
+Gender: Female
+
+Name: en-SG-WayneNeural
+Gender: Male
+
+Name: en-TZ-ElimuNeural
+Gender: Male
+
+Name: en-TZ-ImaniNeural
+Gender: Female
+
+Name: en-US-AnaNeural
+Gender: Female
+
+Name: en-US-AndrewNeural
+Gender: Male
+
+Name: en-US-AriaNeural
+Gender: Female
+
+Name: en-US-AvaNeural
+Gender: Female
+
+Name: en-US-BrianNeural
+Gender: Male
+
+Name: en-US-ChristopherNeural
+Gender: Male
+
+Name: en-US-EmmaNeural
+Gender: Female
+
+Name: en-US-EricNeural
+Gender: Male
+
+Name: en-US-GuyNeural
+Gender: Male
+
+Name: en-US-JennyNeural
+Gender: Female
+
+Name: en-US-MichelleNeural
+Gender: Female
+
+Name: en-US-RogerNeural
+Gender: Male
+
+Name: en-US-SteffanNeural
+Gender: Male
+
+Name: en-ZA-LeahNeural
+Gender: Female
+
+Name: en-ZA-LukeNeural
+Gender: Male
+
+Name: es-AR-ElenaNeural
+Gender: Female
+
+Name: es-AR-TomasNeural
+Gender: Male
+
+Name: es-BO-MarceloNeural
+Gender: Male
+
+Name: es-BO-SofiaNeural
+Gender: Female
+
+Name: es-CL-CatalinaNeural
+Gender: Female
+
+Name: es-CL-LorenzoNeural
+Gender: Male
+
+Name: es-CO-GonzaloNeural
+Gender: Male
+
+Name: es-CO-SalomeNeural
+Gender: Female
+
+Name: es-CR-JuanNeural
+Gender: Male
+
+Name: es-CR-MariaNeural
+Gender: Female
+
+Name: es-CU-BelkysNeural
+Gender: Female
+
+Name: es-CU-ManuelNeural
+Gender: Male
+
+Name: es-DO-EmilioNeural
+Gender: Male
+
+Name: es-DO-RamonaNeural
+Gender: Female
+
+Name: es-EC-AndreaNeural
+Gender: Female
+
+Name: es-EC-LuisNeural
+Gender: Male
+
+Name: es-ES-AlvaroNeural
+Gender: Male
+
+Name: es-ES-ElviraNeural
+Gender: Female
+
+Name: es-ES-XimenaNeural
+Gender: Female
+
+Name: es-GQ-JavierNeural
+Gender: Male
+
+Name: es-GQ-TeresaNeural
+Gender: Female
+
+Name: es-GT-AndresNeural
+Gender: Male
+
+Name: es-GT-MartaNeural
+Gender: Female
+
+Name: es-HN-CarlosNeural
+Gender: Male
+
+Name: es-HN-KarlaNeural
+Gender: Female
+
+Name: es-MX-DaliaNeural
+Gender: Female
+
+Name: es-MX-JorgeNeural
+Gender: Male
+
+Name: es-NI-FedericoNeural
+Gender: Male
+
+Name: es-NI-YolandaNeural
+Gender: Female
+
+Name: es-PA-MargaritaNeural
+Gender: Female
+
+Name: es-PA-RobertoNeural
+Gender: Male
+
+Name: es-PE-AlexNeural
+Gender: Male
+
+Name: es-PE-CamilaNeural
+Gender: Female
+
+Name: es-PR-KarinaNeural
+Gender: Female
+
+Name: es-PR-VictorNeural
+Gender: Male
+
+Name: es-PY-MarioNeural
+Gender: Male
+
+Name: es-PY-TaniaNeural
+Gender: Female
+
+Name: es-SV-LorenaNeural
+Gender: Female
+
+Name: es-SV-RodrigoNeural
+Gender: Male
+
+Name: es-US-AlonsoNeural
+Gender: Male
+
+Name: es-US-PalomaNeural
+Gender: Female
+
+Name: es-UY-MateoNeural
+Gender: Male
+
+Name: es-UY-ValentinaNeural
+Gender: Female
+
+Name: es-VE-PaolaNeural
+Gender: Female
+
+Name: es-VE-SebastianNeural
+Gender: Male
+
+Name: et-EE-AnuNeural
+Gender: Female
+
+Name: et-EE-KertNeural
+Gender: Male
+
+Name: fa-IR-DilaraNeural
+Gender: Female
+
+Name: fa-IR-FaridNeural
+Gender: Male
+
+Name: fi-FI-HarriNeural
+Gender: Male
+
+Name: fi-FI-NooraNeural
+Gender: Female
+
+Name: fil-PH-AngeloNeural
+Gender: Male
+
+Name: fil-PH-BlessicaNeural
+Gender: Female
+
+Name: fr-BE-CharlineNeural
+Gender: Female
+
+Name: fr-BE-GerardNeural
+Gender: Male
+
+Name: fr-CA-AntoineNeural
+Gender: Male
+
+Name: fr-CA-JeanNeural
+Gender: Male
+
+Name: fr-CA-SylvieNeural
+Gender: Female
+
+Name: fr-CA-ThierryNeural
+Gender: Male
+
+Name: fr-CH-ArianeNeural
+Gender: Female
+
+Name: fr-CH-FabriceNeural
+Gender: Male
+
+Name: fr-FR-DeniseNeural
+Gender: Female
+
+Name: fr-FR-EloiseNeural
+Gender: Female
+
+Name: fr-FR-HenriNeural
+Gender: Male
+
+Name: fr-FR-RemyMultilingualNeural
+Gender: Male
+
+Name: fr-FR-VivienneMultilingualNeural
+Gender: Female
+
+Name: ga-IE-ColmNeural
+Gender: Male
+
+Name: ga-IE-OrlaNeural
+Gender: Female
+
+Name: gl-ES-RoiNeural
+Gender: Male
+
+Name: gl-ES-SabelaNeural
+Gender: Female
+
+Name: gu-IN-DhwaniNeural
+Gender: Female
+
+Name: gu-IN-NiranjanNeural
+Gender: Male
+
+Name: he-IL-AvriNeural
+Gender: Male
+
+Name: he-IL-HilaNeural
+Gender: Female
+
+Name: hi-IN-MadhurNeural
+Gender: Male
+
+Name: hi-IN-SwaraNeural
+Gender: Female
+
+Name: hr-HR-GabrijelaNeural
+Gender: Female
+
+Name: hr-HR-SreckoNeural
+Gender: Male
+
+Name: hu-HU-NoemiNeural
+Gender: Female
+
+Name: hu-HU-TamasNeural
+Gender: Male
+
+Name: id-ID-ArdiNeural
+Gender: Male
+
+Name: id-ID-GadisNeural
+Gender: Female
+
+Name: is-IS-GudrunNeural
+Gender: Female
+
+Name: is-IS-GunnarNeural
+Gender: Male
+
+Name: it-IT-DiegoNeural
+Gender: Male
+
+Name: it-IT-ElsaNeural
+Gender: Female
+
+Name: it-IT-GiuseppeNeural
+Gender: Male
+
+Name: it-IT-IsabellaNeural
+Gender: Female
+
+Name: ja-JP-KeitaNeural
+Gender: Male
+
+Name: ja-JP-NanamiNeural
+Gender: Female
+
+Name: jv-ID-DimasNeural
+Gender: Male
+
+Name: jv-ID-SitiNeural
+Gender: Female
+
+Name: ka-GE-EkaNeural
+Gender: Female
+
+Name: ka-GE-GiorgiNeural
+Gender: Male
+
+Name: kk-KZ-AigulNeural
+Gender: Female
+
+Name: kk-KZ-DauletNeural
+Gender: Male
+
+Name: km-KH-PisethNeural
+Gender: Male
+
+Name: km-KH-SreymomNeural
+Gender: Female
+
+Name: kn-IN-GaganNeural
+Gender: Male
+
+Name: kn-IN-SapnaNeural
+Gender: Female
+
+Name: ko-KR-HyunsuNeural
+Gender: Male
+
+Name: ko-KR-InJoonNeural
+Gender: Male
+
+Name: ko-KR-SunHiNeural
+Gender: Female
+
+Name: lo-LA-ChanthavongNeural
+Gender: Male
+
+Name: lo-LA-KeomanyNeural
+Gender: Female
+
+Name: lt-LT-LeonasNeural
+Gender: Male
+
+Name: lt-LT-OnaNeural
+Gender: Female
+
+Name: lv-LV-EveritaNeural
+Gender: Female
+
+Name: lv-LV-NilsNeural
+Gender: Male
+
+Name: mk-MK-AleksandarNeural
+Gender: Male
+
+Name: mk-MK-MarijaNeural
+Gender: Female
+
+Name: ml-IN-MidhunNeural
+Gender: Male
+
+Name: ml-IN-SobhanaNeural
+Gender: Female
+
+Name: mn-MN-BataaNeural
+Gender: Male
+
+Name: mn-MN-YesuiNeural
+Gender: Female
+
+Name: mr-IN-AarohiNeural
+Gender: Female
+
+Name: mr-IN-ManoharNeural
+Gender: Male
+
+Name: ms-MY-OsmanNeural
+Gender: Male
+
+Name: ms-MY-YasminNeural
+Gender: Female
+
+Name: mt-MT-GraceNeural
+Gender: Female
+
+Name: mt-MT-JosephNeural
+Gender: Male
+
+Name: my-MM-NilarNeural
+Gender: Female
+
+Name: my-MM-ThihaNeural
+Gender: Male
+
+Name: nb-NO-FinnNeural
+Gender: Male
+
+Name: nb-NO-PernilleNeural
+Gender: Female
+
+Name: ne-NP-HemkalaNeural
+Gender: Female
+
+Name: ne-NP-SagarNeural
+Gender: Male
+
+Name: nl-BE-ArnaudNeural
+Gender: Male
+
+Name: nl-BE-DenaNeural
+Gender: Female
+
+Name: nl-NL-ColetteNeural
+Gender: Female
+
+Name: nl-NL-FennaNeural
+Gender: Female
+
+Name: nl-NL-MaartenNeural
+Gender: Male
+
+Name: pl-PL-MarekNeural
+Gender: Male
+
+Name: pl-PL-ZofiaNeural
+Gender: Female
+
+Name: ps-AF-GulNawazNeural
+Gender: Male
+
+Name: ps-AF-LatifaNeural
+Gender: Female
+
+Name: pt-BR-AntonioNeural
+Gender: Male
+
+Name: pt-BR-FranciscaNeural
+Gender: Female
+
+Name: pt-BR-ThalitaNeural
+Gender: Female
+
+Name: pt-PT-DuarteNeural
+Gender: Male
+
+Name: pt-PT-RaquelNeural
+Gender: Female
+
+Name: ro-RO-AlinaNeural
+Gender: Female
+
+Name: ro-RO-EmilNeural
+Gender: Male
+
+Name: ru-RU-DmitryNeural
+Gender: Male
+
+Name: ru-RU-SvetlanaNeural
+Gender: Female
+
+Name: si-LK-SameeraNeural
+Gender: Male
+
+Name: si-LK-ThiliniNeural
+Gender: Female
+
+Name: sk-SK-LukasNeural
+Gender: Male
+
+Name: sk-SK-ViktoriaNeural
+Gender: Female
+
+Name: sl-SI-PetraNeural
+Gender: Female
+
+Name: sl-SI-RokNeural
+Gender: Male
+
+Name: so-SO-MuuseNeural
+Gender: Male
+
+Name: so-SO-UbaxNeural
+Gender: Female
+
+Name: sq-AL-AnilaNeural
+Gender: Female
+
+Name: sq-AL-IlirNeural
+Gender: Male
+
+Name: sr-RS-NicholasNeural
+Gender: Male
+
+Name: sr-RS-SophieNeural
+Gender: Female
+
+Name: su-ID-JajangNeural
+Gender: Male
+
+Name: su-ID-TutiNeural
+Gender: Female
+
+Name: sv-SE-MattiasNeural
+Gender: Male
+
+Name: sv-SE-SofieNeural
+Gender: Female
+
+Name: sw-KE-RafikiNeural
+Gender: Male
+
+Name: sw-KE-ZuriNeural
+Gender: Female
+
+Name: sw-TZ-DaudiNeural
+Gender: Male
+
+Name: sw-TZ-RehemaNeural
+Gender: Female
+
+Name: ta-IN-PallaviNeural
+Gender: Female
+
+Name: ta-IN-ValluvarNeural
+Gender: Male
+
+Name: ta-LK-KumarNeural
+Gender: Male
+
+Name: ta-LK-SaranyaNeural
+Gender: Female
+
+Name: ta-MY-KaniNeural
+Gender: Female
+
+Name: ta-MY-SuryaNeural
+Gender: Male
+
+Name: ta-SG-AnbuNeural
+Gender: Male
+
+Name: ta-SG-VenbaNeural
+Gender: Female
+
+Name: te-IN-MohanNeural
+Gender: Male
+
+Name: te-IN-ShrutiNeural
+Gender: Female
+
+Name: th-TH-NiwatNeural
+Gender: Male
+
+Name: th-TH-PremwadeeNeural
+Gender: Female
+
+Name: tr-TR-AhmetNeural
+Gender: Male
+
+Name: tr-TR-EmelNeural
+Gender: Female
+
+Name: uk-UA-OstapNeural
+Gender: Male
+
+Name: uk-UA-PolinaNeural
+Gender: Female
+
+Name: ur-IN-GulNeural
+Gender: Female
+
+Name: ur-IN-SalmanNeural
+Gender: Male
+
+Name: ur-PK-AsadNeural
+Gender: Male
+
+Name: ur-PK-UzmaNeural
+Gender: Female
+
+Name: uz-UZ-MadinaNeural
+Gender: Female
+
+Name: uz-UZ-SardorNeural
+Gender: Male
+
+Name: vi-VN-HoaiMyNeural
+Gender: Female
+
+Name: vi-VN-NamMinhNeural
+Gender: Male
+
+Name: zh-CN-XiaoxiaoNeural
+Gender: Female
+
+Name: zh-CN-XiaoyiNeural
+Gender: Female
+
+Name: zh-CN-YunjianNeural
+Gender: Male
+
+Name: zh-CN-YunxiNeural
+Gender: Male
+
+Name: zh-CN-YunxiaNeural
+Gender: Male
+
+Name: zh-CN-YunyangNeural
+Gender: Male
+
+Name: zh-CN-liaoning-XiaobeiNeural
+Gender: Female
+
+Name: zh-CN-shaanxi-XiaoniNeural
+Gender: Female
+
+Name: zh-HK-HiuGaaiNeural
+Gender: Female
+
+Name: zh-HK-HiuMaanNeural
+Gender: Female
+
+Name: zh-HK-WanLungNeural
+Gender: Male
+
+Name: zh-TW-HsiaoChenNeural
+Gender: Female
+
+Name: zh-TW-HsiaoYuNeural
+Gender: Female
+
+Name: zh-TW-YunJheNeural
+Gender: Male
+
+Name: zu-ZA-ThandoNeural
+Gender: Female
+
+Name: zu-ZA-ThembaNeural
+Gender: Male

sites/docs/README.md

@@ -0,0 +1,16 @@
+---
+home: true
+heroImage: /hero.png
+actions:
+  - text: Get Started →
+    link: /guide/
+    type: primary
+features:
+  - title: Multilingual
+    details: Supports video scripts in both Chinese and English; offers multiple voice synthesis options.
+  - title: Maintainability
+    details: Complete MVC architecture with clear code structure, easy to maintain, supports both API and Web interface.
+  - title: Multi-Model Support
+    details: Supports integration with multiple models including OpenAI, moonshot, Azure, gpt4free, one-api, Tongyi Qianwen, Google Gemini, Ollama, and others.
+footer: MIT Licensed | Copyright © 2024-present MoneyPrinterTurbo
+---

sites/docs/guide/README.md

@@ -0,0 +1,134 @@
+## Installation & Deployment 📥
+
+Simply provide a <b>topic</b> or <b>keyword</b> for a video, and it will automatically generate the video copy, video
+materials, video subtitles, and video background music before synthesizing a high-definition short video.
+
+### WebUI
+
+![](/webui-en.jpg)
+
+### API Interface
+
+![](/api.jpg)
+
+- Try to avoid using **Chinese paths** to prevent unpredictable issues
+- Ensure your **network** is stable, meaning you can access foreign websites normally
+
+#### ① Clone the Project
+
+```shell
+git clone https://github.com/harry0703/MoneyPrinterTurbo.git
+```
+
+#### ② Modify the Configuration File
+
+- Copy the `config.example.toml` file and rename it to `config.toml`
+- Follow the instructions in the `config.toml` file to configure `pexels_api_keys` and `llm_provider`, and according to
+  the llm_provider's service provider, set up the corresponding API Key
+
+#### ③ Configure Large Language Models (LLM)
+
+- To use `GPT-4.0` or `GPT-3.5`, you need an `API Key` from `OpenAI`. If you don't have one, you can set `llm_provider`
+  to `g4f` (a free-to-use GPT library https://github.com/xtekky/gpt4free)
+
+### Docker Deployment 🐳
+
+#### ① Launch the Docker Container
+
+If you haven't installed Docker, please install it first https://www.docker.com/products/docker-desktop/
+If you are using a Windows system, please refer to Microsoft's documentation:
+
+1. https://learn.microsoft.com/en-us/windows/wsl/install
+2. https://learn.microsoft.com/en-us/windows/wsl/tutorials/wsl-containers
+
+```shell
+cd MoneyPrinterTurbo
+docker-compose up
+```
+
+#### ② Access the Web Interface
+
+Open your browser and visit http://0.0.0.0:8501
+
+#### ③ Access the API Interface
+
+Open your browser and visit http://0.0.0.0:8080/docs Or http://0.0.0.0:8080/redoc
+
+### Manual Deployment 📦
+
+#### ① Create a Python Virtual Environment
+
+It is recommended to create a Python virtual environment
+using [conda](https://conda.io/projects/conda/en/latest/user-guide/install/index.html)
+
+```shell
+git clone https://github.com/harry0703/MoneyPrinterTurbo.git
+cd MoneyPrinterTurbo
+conda create -n MoneyPrinterTurbo python=3.10
+conda activate MoneyPrinterTurbo
+pip install -r requirements.txt
+```
+
+#### ② Install ImageMagick
+
+###### Windows:
+
+- Download https://imagemagick.org/archive/binaries/ImageMagick-7.1.1-29-Q16-x64-static.exe
+- Install the downloaded ImageMagick, **do not change the installation path**
+- Modify the `config.toml` configuration file, set `imagemagick_path` to your actual installation path (if you didn't
+  change the path during installation, just uncomment it)
+
+###### MacOS:
+
+```shell
+brew install imagemagick
+```
+
+###### Ubuntu
+
+```shell
+sudo apt-get install imagemagick
+```
+
+###### CentOS
+
+```shell
+sudo yum install ImageMagick
+```
+
+#### ③ Launch the Web Interface 🌐
+
+Note that you need to execute the following commands in the `root directory` of the MoneyPrinterTurbo project
+
+###### Windows
+
+```bat
+conda activate MoneyPrinterTurbo
+webui.bat
+```
+
+###### MacOS or Linux
+
+```shell
+conda activate MoneyPrinterTurbo
+sh webui.sh
+```
+
+After launching, the browser will open automatically
+
+#### ④ Launch the API Service 🚀
+
+```shell
+python main.py
+```
+
+After launching, you can view the `API documentation` at http://127.0.0.1:8080/docs and directly test the interface
+online for a quick experience.
+
+## License 📝
+
+Click to view the [`LICENSE`](LICENSE) file
+
+## Star History
+
+[![Star History Chart](https://api.star-history.com/svg?repos=harry0703/MoneyPrinterTurbo&type=Date)](https://star-history.com/#harry0703/MoneyPrinterTurbo&Date)

sites/docs/guide/background-music.md

@@ -0,0 +1,5 @@
+## Background Music 🎵
+
+Background music for videos is located in the project's `resource/songs` directory.
+> The current project includes some default music from YouTube videos. If there are copyright issues, please delete
+> them.

sites/docs/guide/faq.md

@@ -0,0 +1,70 @@
+## Common Questions 🤔
+
+### ❓How to Use the Free OpenAI GPT-3.5 Model?
+
+[OpenAI has announced that ChatGPT with 3.5 is now free](https://openai.com/blog/start-using-chatgpt-instantly), and
+developers have wrapped it into an API for direct usage.
+
+**Ensure you have Docker installed and running**. Execute the following command to start the Docker service:
+
+```shell
+docker run -p 3040:3040 missuo/freegpt35
+```
+
+Once successfully started, modify the `config.toml` configuration as follows:
+
+- Set `llm_provider` to `openai`
+- Fill in `openai_api_key` with any value, for example, '123456'
+- Change `openai_base_url` to `http://localhost:3040/v1/`
+- Set `openai_model_name` to `gpt-3.5-turbo`
+
+### ❓RuntimeError: No ffmpeg exe could be found
+
+Normally, ffmpeg will be automatically downloaded and detected.
+However, if your environment has issues preventing automatic downloads, you may encounter the following error:
+
+```
+RuntimeError: No ffmpeg exe could be found.
+Install ffmpeg on your system, or set the IMAGEIO_FFMPEG_EXE environment variable.
+```
+
+In this case, you can download ffmpeg from https://www.gyan.dev/ffmpeg/builds/, unzip it, and set `ffmpeg_path` to your
+actual installation path.
+
+```toml
+[app]
+# Please set according to your actual path, note that Windows path separators are \\
+ffmpeg_path = "C:\\Users\\harry\\Downloads\\ffmpeg.exe"
+```
+
+### ❓Error generating audio or downloading videos
+
+[issue 56](https://github.com/harry0703/MoneyPrinterTurbo/issues/56)
+
+```
+failed to generate audio, maybe the network is not available. 
+if you are in China, please use a VPN.
+```
+
+[issue 44](https://github.com/harry0703/MoneyPrinterTurbo/issues/44)
+
+```
+failed to download videos, maybe the network is not available. 
+if you are in China, please use a VPN.
+```
+
+This is likely due to network issues preventing access to foreign services. Please use a VPN to resolve this.
+
+### ❓ImageMagick is not installed on your computer
+
+[issue 33](https://github.com/harry0703/MoneyPrinterTurbo/issues/33)
+
+1. Follow the `example configuration` provided `download address` to
+   install https://imagemagick.org/archive/binaries/ImageMagick-7.1.1-30-Q16-x64-static.exe, using the static library
+2. Do not install in a path with Chinese characters to avoid unpredictable issues
+
+[issue 54](https://github.com/harry0703/MoneyPrinterTurbo/issues/54#issuecomment-2017842022)
+
+For Linux systems, you can manually install it, refer to https://cn.linux-console.net/?p=16978
+
+Thanks to [@wangwenqiao666](https://github.com/wangwenqiao666) for their research and exploration

sites/docs/guide/features.md

@@ -0,0 +1,34 @@
+## Features 🎯
+
+- [x] Complete **MVC architecture**, **clearly structured** code, easy to maintain, supports both `API`
+  and `Web interface`
+- [x] Supports **AI-generated** video copy, as well as **customized copy**
+- [x] Supports various **high-definition video** sizes
+    - [x] Portrait 9:16, `1080x1920`
+    - [x] Landscape 16:9, `1920x1080`
+- [x] Supports **batch video generation**, allowing the creation of multiple videos at once, then selecting the most
+  satisfactory one
+- [x] Supports setting the **duration of video clips**, facilitating adjustments to material switching frequency
+- [x] Supports video copy in both **Chinese** and **English**
+- [x] Supports **multiple voice** synthesis
+- [x] Supports **subtitle generation**, with adjustable `font`, `position`, `color`, `size`, and also
+  supports `subtitle outlining`
+- [x] Supports **background music**, either random or specified music files, with adjustable `background music volume`
+- [x] Video material sources are **high-definition** and **royalty-free**
+- [x] Supports integration with various models such as **OpenAI**, **moonshot**, **Azure**, **gpt4free**, **one-api**,
+  **qianwen**, **Google Gemini**, **Ollama** and more
+
+❓[How to Use the Free OpenAI GPT-3.5 Model?](https://github.com/harry0703/MoneyPrinterTurbo/blob/main/README-en.md#common-questions-)
+
+### Future Plans 📅
+
+- [ ] Introduce support for GPT-SoVITS dubbing
+- [ ] Enhance voice synthesis with large models for a more natural and emotionally resonant voice output
+- [ ] Incorporate video transition effects to ensure a smoother viewing experience
+- [ ] Improve the relevance of video content
+- [ ] Add options for video length: short, medium, long
+- [ ] Package the application into a one-click launch bundle for Windows and macOS for ease of use
+- [ ] Enable the use of custom materials
+- [ ] Offer voiceover and background music options with real-time preview
+- [ ] Support a wider range of voice synthesis providers, such as OpenAI TTS, Azure TTS
+- [ ] Automate the upload process to the YouTube platform

sites/docs/guide/feedback.md

@@ -0,0 +1,4 @@
+## Feedback & Suggestions 📢
+
+- You can submit an [issue](https://github.com/harry0703/MoneyPrinterTurbo/issues) or
+  a [pull request](https://github.com/harry0703/MoneyPrinterTurbo/pulls).

sites/docs/guide/reference-project.md

@@ -0,0 +1,4 @@
+## Reference Projects 📚
+
+This project is based on https://github.com/FujiwaraChoki/MoneyPrinter and has been refactored with a lot of
+optimizations and added functionalities. Thanks to the original author for their spirit of open source.

sites/docs/guide/speech-synthesis.md

@@ -0,0 +1,3 @@
+## Voice Synthesis 🗣
+
+A list of all supported voices can be viewed here: [Voice List](/voice-list.txt)

sites/docs/guide/subtitle-font.md

@@ -0,0 +1,4 @@
+## Subtitle Fonts 🅰
+
+Fonts for rendering video subtitles are located in the project's `resource/fonts` directory, and you can also add your
+own fonts.

sites/docs/guide/subtitle-generation.md

@@ -0,0 +1,15 @@
+## Subtitle Generation 📜
+
+Currently, there are 2 ways to generate subtitles:
+
+- edge: Faster generation speed, better performance, no specific requirements for computer configuration, but the
+  quality may be unstable
+- whisper: Slower generation speed, poorer performance, specific requirements for computer configuration, but more
+  reliable quality
+
+You can switch between them by modifying the `subtitle_provider` in the `config.toml` configuration file
+
+It is recommended to use `edge` mode, and switch to `whisper` mode if the quality of the subtitles generated is not
+satisfactory.
+
+> If left blank, it means no subtitles will be generated.

sites/docs/guide/video-demonstration.md

@@ -0,0 +1,35 @@
+## Video Demos 📺
+
+### Portrait 9:16
+
+<table>
+<thead>
+<tr>
+<th align="center"><g-emoji class="g-emoji" alias="arrow_forward">▶️</g-emoji> How to Add Fun to Your Life </th>
+<th align="center"><g-emoji class="g-emoji" alias="arrow_forward">▶️</g-emoji> What is the Meaning of Life</th>
+</tr>
+</thead>
+<tbody>
+<tr>
+<td align="center"><video src="https://github.com/harry0703/MoneyPrinterTurbo/assets/4928832/a84d33d5-27a2-4aba-8fd0-9fb2bd91c6a6"></video></td>
+<td align="center"><video src="https://github.com/harry0703/MoneyPrinterTurbo/assets/4928832/112c9564-d52b-4472-99ad-970b75f66476"></video></td>
+</tr>
+</tbody>
+</table>
+
+### Landscape 16:9
+
+<table>
+<thead>
+<tr>
+<th align="center"><g-emoji class="g-emoji" alias="arrow_forward">▶️</g-emoji> What is the Meaning of Life</th>
+<th align="center"><g-emoji class="g-emoji" alias="arrow_forward">▶️</g-emoji> Why Exercise</th>
+</tr>
+</thead>
+<tbody>
+<tr>
+<td align="center"><video src="https://github.com/harry0703/MoneyPrinterTurbo/assets/4928832/346ebb15-c55f-47a9-a653-114f08bb8073"></video></td>
+<td align="center"><video src="https://github.com/harry0703/MoneyPrinterTurbo/assets/4928832/271f2fae-8283-44a0-8aa0-0ed8f9a6fa87"></video></td>
+</tr>
+</tbody>
+</table>

sites/docs/zh/README.md

@@ -0,0 +1,16 @@
+---
+home: true
+heroImage: /hero.png
+actions:
+  - text: 快速上手 →
+    link: /zh/guide/
+    type: primary
+features:
+  - title: 多语言
+    details: 支持 中文 和 英文 视频文案；支持 多种语音 合成。
+  - title: 可维护性
+    details: 完整的 MVC架构，代码 结构清晰，易于维护，支持 API 和 Web界面。
+  - title: 多模型支持
+    details: 支持 OpenAI、moonshot、Azure、gpt4free、one-api、通义千问、Google Gemini、Ollama 等多种模型接入。
+footer: MIT Licensed | Copyright © 2024-present MoneyPrinterTurbo
+---

sites/docs/zh/guide/README.md

@@ -0,0 +1,157 @@
+## 快速开始 🚀
+
+<br>
+只需提供一个视频 <b>主题</b> 或 <b>关键词</b> ，就可以全自动生成视频文案、视频素材、视频字幕、视频背景音乐，然后合成一个高清的短视频。
+<br>
+
+<h4>Web界面</h4>
+
+![](/webui.jpg)
+
+<h4>API界面</h4>
+
+![](/api.jpg)
+
+下载一键启动包，解压直接使用
+
+### Windows
+
+- 百度网盘: https://pan.baidu.com/s/1bpGjgQVE5sADZRn3A6F87w?pwd=xt16 提取码: xt16
+
+下载后，建议先**双击执行** `update.bat` 更新到**最新代码**，然后双击 `start.bat` 启动 Web 界面
+
+### 其他系统
+
+还没有制作一键启动包，看下面的 **安装部署** 部分，建议使用 **docker** 部署，更加方便。
+
+## 安装部署 📥
+
+### 前提条件
+
+- 尽量不要使用 **中文路径**，避免出现一些无法预料的问题
+- 请确保你的 **网络** 是正常的，VPN 需要打开`全局流量`模式
+
+#### ① 克隆代码
+
+```shell
+git clone https://github.com/harry0703/MoneyPrinterTurbo.git
+```
+
+#### ② 修改配置文件
+
+- 将 `config.example.toml` 文件复制一份，命名为 `config.toml`
+- 按照 `config.toml` 文件中的说明，配置好 `pexels_api_keys` 和 `llm_provider`，并根据 llm_provider 对应的服务商，配置相关的
+  API Key
+
+#### ③ 配置大模型(LLM)
+
+- 如果要使用 `GPT-4.0` 或 `GPT-3.5`，需要有 `OpenAI` 的 `API Key`，如果没有，可以将 `llm_provider` 设置为 `g4f` (
+  一个免费使用 GPT 的开源库 https://github.com/xtekky/gpt4free ，但是该免费的服务，稳定性较差，有时候可以用，有时候用不了)
+- 或者可以使用到 [月之暗面](https://platform.moonshot.cn/console/api-keys) 申请。注册就送
+  15 元体验金，可以对话 1500 次左右。然后设置 `llm_provider="moonshot"` 和 `moonshot_api_key`
+- 也可以使用 通义千问，具体请看配置文件里面的注释说明
+
+### Docker 部署 🐳
+
+#### ① 启动 Docker
+
+如果未安装 Docker，请先安装 https://www.docker.com/products/docker-desktop/
+
+如果是 Windows 系统，请参考微软的文档：
+
+1. https://learn.microsoft.com/zh-cn/windows/wsl/install
+2. https://learn.microsoft.com/zh-cn/windows/wsl/tutorials/wsl-containers
+
+```shell
+cd MoneyPrinterTurbo
+docker-compose up
+```
+
+#### ② 访问 Web 界面
+
+打开浏览器，访问 http://0.0.0.0:8501
+
+#### ③ 访问 API 文档
+
+打开浏览器，访问 http://0.0.0.0:8080/docs 或者 http://0.0.0.0:8080/redoc
+
+### 手动部署 📦
+
+> 视频教程
+
+- 完整的使用演示：https://v.douyin.com/iFhnwsKY/
+- 如何在 Windows 上部署：https://v.douyin.com/iFyjoW3M
+
+#### ① 创建虚拟环境
+
+建议使用 [conda](https://conda.io/projects/conda/en/latest/user-guide/install/index.html) 创建 python 虚拟环境
+
+```shell
+git clone https://github.com/harry0703/MoneyPrinterTurbo.git
+cd MoneyPrinterTurbo
+conda create -n MoneyPrinterTurbo python=3.10
+conda activate MoneyPrinterTurbo
+pip install -r requirements.txt
+```
+
+#### ② 安装好 ImageMagick
+
+###### Windows:
+
+- 下载 https://imagemagick.org/archive/binaries/ImageMagick-7.1.1-30-Q16-x64-static.exe
+- 安装下载好的 ImageMagick，注意不要修改安装路径
+- 修改 `配置文件 config.toml` 中的 `imagemagick_path` 为你的实际安装路径（如果安装的时候没有修改路径，直接取消注释即可）
+
+###### MacOS:
+
+```shell
+brew install imagemagick
+```
+
+###### Ubuntu
+
+```shell
+sudo apt-get install imagemagick
+```
+
+###### CentOS
+
+```shell
+sudo yum install ImageMagick
+```
+
+#### ③ 启动 Web 界面 🌐
+
+注意需要到 MoneyPrinterTurbo 项目 `根目录` 下执行以下命令
+
+###### Windows
+
+```bat
+conda activate MoneyPrinterTurbo
+webui.bat
+```
+
+###### MacOS or Linux
+
+```shell
+conda activate MoneyPrinterTurbo
+sh webui.sh
+```
+
+启动后，会自动打开浏览器
+
+#### ④ 启动 API 服务 🚀
+
+```shell
+python main.py
+```
+
+启动后，可以查看 `API文档` http://127.0.0.1:8080/docs 或者 http://127.0.0.1:8080/redoc 直接在线调试接口，快速体验。
+
+## 许可证 📝
+
+点击查看 [`LICENSE`](LICENSE) 文件
+
+## Star History
+
+[![Star History Chart](https://api.star-history.com/svg?repos=harry0703/MoneyPrinterTurbo&type=Date)](https://star-history.com/#harry0703/MoneyPrinterTurbo&Date)

sites/docs/zh/guide/background-music.md

@@ -0,0 +1,4 @@
+## 背景音乐 🎵
+
+用于视频的背景音乐，位于项目的 `resource/songs` 目录下。
+> 当前项目里面放了一些默认的音乐，来自于 YouTube 视频，如有侵权，请删除。

sites/docs/zh/guide/configuration-requirements.md

@@ -0,0 +1,4 @@
+## 配置要求 📦
+
+- 建议最低 CPU 4核或以上，内存 8G 或以上，显卡非必须
+- Windows 10 或 MacOS 11.0 以上系统

sites/docs/zh/guide/faq.md

@@ -0,0 +1,123 @@
+## 常见问题 🤔
+
+### ❓如何使用免费的OpenAI GPT-3.5模型?
+
+[OpenAI宣布ChatGPT里面3.5已经免费了](https://openai.com/blog/start-using-chatgpt-instantly)，有开发者将其封装成了API，可以直接调用
+
+**确保你安装和启动了docker服务**，执行以下命令启动docker服务
+
+```shell
+docker run -p 3040:3040 missuo/freegpt35
+```
+
+启动成功后，修改 `config.toml` 中的配置
+
+- `llm_provider` 设置为 `openai`
+- `openai_api_key` 随便填写一个即可，比如 '123456'
+- `openai_base_url` 改为 `http://localhost:3040/v1/`
+- `openai_model_name` 改为 `gpt-3.5-turbo`
+
+### ❓AttributeError: 'str' object has no attribute 'choices'`
+
+这个问题是由于 OpenAI 或者其他 LLM，没有返回正确的回复导致的。
+
+大概率是网络原因， 使用 **VPN**，或者设置 `openai_base_url` 为你的代理 ，应该就可以解决了。
+
+### ❓RuntimeError: No ffmpeg exe could be found
+
+通常情况下，ffmpeg 会被自动下载，并且会被自动检测到。
+但是如果你的环境有问题，无法自动下载，可能会遇到如下错误：
+
+```
+RuntimeError: No ffmpeg exe could be found.
+Install ffmpeg on your system, or set the IMAGEIO_FFMPEG_EXE environment variable.
+```
+
+此时你可以从 https://www.gyan.dev/ffmpeg/builds/ 下载ffmpeg，解压后，设置 `ffmpeg_path` 为你的实际安装路径即可。
+
+```toml
+[app]
+# 请根据你的实际路径设置，注意 Windows 路径分隔符为 \\
+ffmpeg_path = "C:\\Users\\harry\\Downloads\\ffmpeg.exe"
+```
+
+### ❓生成音频时报错或下载视频报错
+
+[issue 56](https://github.com/harry0703/MoneyPrinterTurbo/issues/56)
+
+```
+failed to generate audio, maybe the network is not available. 
+if you are in China, please use a VPN.
+```
+
+[issue 44](https://github.com/harry0703/MoneyPrinterTurbo/issues/44)
+
+```
+failed to download videos, maybe the network is not available. 
+if you are in China, please use a VPN.
+```
+
+这个大概率是网络原因，无法访问境外的服务，请使用VPN解决。
+
+### ❓ImageMagick is not installed on your computer
+
+[issue 33](https://github.com/harry0703/MoneyPrinterTurbo/issues/33)
+
+1. 按照 `示例配置` 里面提供的 `下载地址`
+   ，安装 https://imagemagick.org/archive/binaries/ImageMagick-7.1.1-29-Q16-x64-static.exe, 用静态库
+2. 不要安装在中文路径里面，避免出现一些无法预料的问题
+
+[issue 54](https://github.com/harry0703/MoneyPrinterTurbo/issues/54#issuecomment-2017842022)
+
+如果是linux系统，可以手动安装，参考 https://cn.linux-console.net/?p=16978
+
+感谢 [@wangwenqiao666](https://github.com/wangwenqiao666)的研究探索
+
+### ❓ImageMagick的安全策略阻止了与临时文件@/tmp/tmpur5hyyto.txt相关的操作
+
+[issue 92](https://github.com/harry0703/MoneyPrinterTurbo/issues/92)
+
+可以在ImageMagick的配置文件policy.xml中找到这些策略。
+这个文件通常位于 /etc/ImageMagick-`X`/ 或 ImageMagick 安装目录的类似位置。
+修改包含`pattern="@"`的条目，将`rights="none"`更改为`rights="read|write"`以允许对文件的读写操作。
+
+感谢 [@chenhengzh](https://github.com/chenhengzh)的研究探索
+
+### ❓OSError: [Errno 24] Too many open files
+
+[issue 100](https://github.com/harry0703/MoneyPrinterTurbo/issues/100)
+
+这个问题是由于系统打开文件数限制导致的，可以通过修改系统的文件打开数限制来解决。
+
+查看当前限制
+
+```shell
+ulimit -n
+```
+
+如果过低，可以调高一些，比如
+
+```shell
+ulimit -n 10240
+```
+
+### ❓AttributeError: module 'PIL.Image' has no attribute 'ANTIALIAS'
+
+[issue 101](https://github.com/harry0703/MoneyPrinterTurbo/issues/101),
+[issue 83](https://github.com/harry0703/MoneyPrinterTurbo/issues/83),
+[issue 70](https://github.com/harry0703/MoneyPrinterTurbo/issues/70)
+
+先看下当前的 Pillow 版本是多少
+
+```shell
+pip list |grep Pillow
+```
+
+如果是 10.x 的版本，可以尝试下降级看看，有用户反馈降级后正常
+
+```shell
+pip uninstall Pillow
+pip install Pillow==9.5.0
+# 或者降级到 8.4.0
+pip install Pillow==8.4.0
+```

sites/docs/zh/guide/features.md

@@ -0,0 +1,31 @@
+## 功能特性 🎯
+
+- [x] 完整的 **MVC架构**，代码 **结构清晰**，易于维护，支持 `API` 和 `Web界面`
+- [x] 支持视频文案 **AI自动生成**，也可以**自定义文案**
+- [x] 支持多种 **高清视频** 尺寸
+    - [x] 竖屏 9:16，`1080x1920`
+    - [x] 横屏 16:9，`1920x1080`
+- [x] 支持 **批量视频生成**，可以一次生成多个视频，然后选择一个最满意的
+- [x] 支持 **视频片段时长**设置，方便调节素材切换频率
+- [x] 支持 **中文** 和 **英文** 视频文案
+- [x] 支持 **多种语音** 合成
+- [x] 支持 **字幕生成**，可以调整 `字体`、`位置`、`颜色`、`大小`，同时支持`字幕描边`设置
+- [x] 支持 **背景音乐**，随机或者指定音乐文件，可设置`背景音乐音量`
+- [x] 视频素材来源 **高清**，而且 **无版权**
+- [x] 支持 **OpenAI**、**moonshot**、**Azure**、**gpt4free**、**one-api**、**通义千问**、**Google Gemini**、**Ollama** 等多种模型接入
+
+  ❓[如何使用免费的 **OpenAI GPT-3.5
+  ** 模型?](https://github.com/harry0703/MoneyPrinterTurbo?tab=readme-ov-file#%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98-)
+
+  ### 后期计划 📅
+
+- [ ] GPT-SoVITS 配音支持
+- [ ] 优化语音合成，利用大模型，使其合成的声音，更加自然，情绪更加丰富
+- [ ] 增加视频转场效果，使其看起来更加的流畅
+- [ ] 增加更多视频素材来源，优化视频素材和文案的匹配度
+- [ ] 增加视频长度选项：短、中、长
+- [ ] 增加免费网络代理，让访问OpenAI和素材下载不再受限
+- [ ] 可以使用自己的素材
+- [ ] 朗读声音和背景音乐，提供实时试听
+- [ ] 支持更多的语音合成服务商，比如 OpenAI TTS
+- [ ] 自动上传到YouTube平台

sites/docs/zh/guide/feedback.md

@@ -0,0 +1,4 @@
+## 反馈建议 📢
+
+- 可以提交 [issue](https://github.com/harry0703/MoneyPrinterTurbo/issues)
+  或者 [pull request](https://github.com/harry0703/MoneyPrinterTurbo/pulls)。

sites/docs/zh/guide/reference-project.md

@@ -0,0 +1,4 @@
+## 参考项目 📚
+
+该项目基于 https://github.com/FujiwaraChoki/MoneyPrinter 重构而来，做了大量的优化，增加了更多的功能。
+感谢原作者的开源精神。

sites/docs/zh/guide/special-thanks.md

@@ -0,0 +1,9 @@
+## 特别感谢 🙏
+
+由于该项目的 **部署** 和 **使用**，对于一些小白用户来说，还是 **有一定的门槛**，在此特别感谢
+**录咖（AI智能 多媒体服务平台）** 网站基于该项目，提供的免费`AI视频生成器`服务，可以不用部署，直接在线使用，非常方便。
+
+- 中文版：https://reccloud.cn
+- 英文版：https://reccloud.com
+
+![](/reccloud.cn.jpg)

sites/docs/zh/guide/speech-synthesis.md

@@ -0,0 +1,5 @@
+## 语音合成 🗣
+
+所有支持的声音列表，可以查看：[声音列表](/voice-list.txt)
+
+2024-04-16 v1.1.2 新增了9种Azure的语音合成声音，需要配置API KEY，该声音合成的更加真实。

sites/docs/zh/guide/subtitle-font.md

@@ -0,0 +1,3 @@
+## 字幕字体 🅰
+
+用于视频字幕的渲染，位于项目的 `resource/fonts` 目录下，你也可以放进去自己的字体。

sites/docs/zh/guide/subtitle-generation.md

@@ -0,0 +1,36 @@
+## 字幕生成 📜
+
+当前支持2种字幕生成方式：
+
+- **edge**: 生成`速度快`，性能更好，对电脑配置没有要求，但是质量可能不稳定
+- **whisper**: 生成`速度慢`，性能较差，对电脑配置有一定要求，但是`质量更可靠`。
+
+可以修改 `config.toml` 配置文件中的 `subtitle_provider` 进行切换
+
+建议使用 `edge` 模式，如果生成的字幕质量不好，再切换到 `whisper` 模式
+
+> 注意：
+
+1. whisper 模式下需要到 HuggingFace 下载一个模型文件，大约 3GB 左右，请确保网络通畅
+2. 如果留空，表示不生成字幕。
+
+> 由于国内无法访问 HuggingFace，可以使用以下方法下载 `whisper-large-v3` 的模型文件
+
+下载地址：
+
+- 百度网盘: https://pan.baidu.com/s/11h3Q6tsDtjQKTjUu3sc5cA?pwd=xjs9
+- 夸克网盘：https://pan.quark.cn/s/3ee3d991d64b
+
+模型下载后解压，整个目录放到 `.\MoneyPrinterTurbo\models` 里面，
+最终的文件路径应该是这样: `.\MoneyPrinterTurbo\models\whisper-large-v3`
+
+```
+MoneyPrinterTurbo  
+  ├─models
+  │   └─whisper-large-v3
+  │          config.json
+  │          model.bin
+  │          preprocessor_config.json
+  │          tokenizer.json
+  │          vocabulary.json
+```

sites/docs/zh/guide/thanks-for-sponsoring.md

@@ -0,0 +1,7 @@
+## 感谢赞助 🙏
+
+感谢佐糖 https://picwish.cn 对该项目的支持和赞助，使得该项目能够持续的更新和维护。
+
+佐糖专注于**图像处理领域**，提供丰富的**图像处理工具**，将复杂操作极致简化，真正实现让图像处理更简单。
+
+![picwish.jpg](/picwish.jpg)

sites/docs/zh/guide/video-demonstration.md

@@ -0,0 +1,37 @@
+## 视频演示 📺
+
+### 竖屏 9:16
+
+<table>
+<thead>
+<tr>
+<th align="center"><g-emoji class="g-emoji" alias="arrow_forward">▶️</g-emoji> 《如何增加生活的乐趣》</th>
+<th align="center"><g-emoji class="g-emoji" alias="arrow_forward">▶️</g-emoji> 《金钱的作用》<br>更真实的合成声音</th>
+<th align="center"><g-emoji class="g-emoji" alias="arrow_forward">▶️</g-emoji> 《生命的意义是什么》</th>
+</tr>
+</thead>
+<tbody>
+<tr>
+<td align="center"><video src="https://github.com/harry0703/MoneyPrinterTurbo/assets/4928832/a84d33d5-27a2-4aba-8fd0-9fb2bd91c6a6"></video></td>
+<td align="center"><video src="https://github.com/harry0703/MoneyPrinterTurbo/assets/4928832/af2f3b0b-002e-49fe-b161-18ba91c055e8"></video></td>
+<td align="center"><video src="https://github.com/harry0703/MoneyPrinterTurbo/assets/4928832/112c9564-d52b-4472-99ad-970b75f66476"></video></td>
+</tr>
+</tbody>
+</table>
+
+### 横屏 16:9
+
+<table>
+<thead>
+<tr>
+<th align="center"><g-emoji class="g-emoji" alias="arrow_forward">▶️</g-emoji>《生命的意义是什么》</th>
+<th align="center"><g-emoji class="g-emoji" alias="arrow_forward">▶️</g-emoji>《为什么要运动》</th>
+</tr>
+</thead>
+<tbody>
+<tr>
+<td align="center"><video src="https://github.com/harry0703/MoneyPrinterTurbo/assets/4928832/346ebb15-c55f-47a9-a653-114f08bb8073"></video></td>
+<td align="center"><video src="https://github.com/harry0703/MoneyPrinterTurbo/assets/4928832/271f2fae-8283-44a0-8aa0-0ed8f9a6fa87"></video></td>
+</tr>
+</tbody>
+</table>

sites/package.json

@@ -0,0 +1,24 @@
+{
+  "name": "MoneyPrinterTurbo",
+  "version": "1.1.2",
+  "description": "利用AI大模型，一键生成高清短视频 Generate short videos with one click using AI LLM.",
+  "main": "index.js",
+  "repository": "https://github.com/harry0703/MoneyPrinterTurbo",
+  "author": "harry0703",
+  "license": "MIT",
+  "devDependencies": {
+    "@vue/tsconfig": "^0.1.3",
+    "@vuepress/bundler-vite": "2.0.0-rc.9",
+    "@vuepress/theme-default": "2.0.0-rc.25",
+    "gh-pages": "^6.1.1",
+    "vue": "^3.4.23",
+    "vue-router": "^4.3.1",
+    "vuepress": "2.0.0-rc.9"
+  },
+  "scripts": {
+    "docs:dev": "vuepress dev docs",
+    "docs:build": "vuepress build docs",
+    "predeploy": "pnpm docs:build",
+    "deploy": "gh-pages -d docs/.vuepress/dist"
+  }
+}

sites/tsconfig.config.json

@@ -0,0 +1,14 @@
+{
+    "extends": "@vue/tsconfig/tsconfig.node.json",
+    "include": [
+        "vite.config.*",
+        "vitest.config.*",
+        "cypress.config.*"
+    ],
+    "compilerOptions": {
+        "composite": true,
+        "types": [
+            "node"
+        ]
+    }
+}

sites/tsconfig.json

@@ -0,0 +1,24 @@
+{
+  "extends": "@vue/tsconfig/tsconfig.web.json",
+  "module": "esnext",
+  "include": [
+    "env.d.ts",
+    "src/**/*",
+    "src/**/*.vue",
+    "docs/.vuepress/*.ts"
+  ],
+  "compilerOptions": {
+    "baseUrl": ".",
+    "isolatedModules": true,
+    "paths": {
+      "@/*": [
+        "./src/*"
+      ]
+    }
+  },
+  "references": [
+    {
+      "path": "./tsconfig.config.json"
+    }
+  ]
+}

webui/Main.py

@@ -1,6 +1,5 @@
-import sys
 import os
-import time
+import sys
 
 # Add the root directory of the project to the system path to allow importing modules from the project
 root_dir = os.path.dirname(os.path.dirname(os.path.realpath(__file__)))
@@ -10,30 +9,33 @@ if root_dir not in sys.path:
     print(sys.path)
     print("")
 
-import streamlit as st
-
 import os
-from uuid import uuid4
 import platform
-import streamlit.components.v1 as components
+from uuid import uuid4
+
+import streamlit as st
 from loguru import logger
 
-st.set_page_config(page_title="MoneyPrinterTurbo",
-                   page_icon="🤖",
-                   layout="wide",
-                   initial_sidebar_state="auto",
-                   menu_items={
-                       'Report a bug': "https://github.com/harry0703/MoneyPrinterTurbo/issues",
-                       'About': "# MoneyPrinterTurbo\nSimply provide a topic or keyword for a video, and it will "
-                                "automatically generate the video copy, video materials, video subtitles, "
-                                "and video background music before synthesizing a high-definition short "
-                                "video.\n\nhttps://github.com/harry0703/MoneyPrinterTurbo"
-                   })
+st.set_page_config(
+    page_title="MoneyPrinterTurbo",
+    page_icon="🤖",
+    layout="wide",
+    initial_sidebar_state="auto",
+    menu_items={
+        "Report a bug": "https://github.com/harry0703/MoneyPrinterTurbo/issues",
+        "About": "# MoneyPrinterTurbo\nSimply provide a topic or keyword for a video, and it will "
+        "automatically generate the video copy, video materials, video subtitles, "
+        "and video background music before synthesizing a high-definition short "
+        "video.\n\nhttps://github.com/harry0703/MoneyPrinterTurbo",
+    },
+)
 
-from app.models.schema import VideoParams, VideoAspect, VideoConcatMode
-from app.services import task as tm, llm, voice
-from app.utils import utils
 from app.config import config
+from app.models.const import FILE_TYPE_IMAGES, FILE_TYPE_VIDEOS
+from app.models.schema import MaterialInfo, VideoAspect, VideoConcatMode, VideoParams
+from app.services import llm, voice
+from app.services import task as tm
+from app.utils import utils
 
 hide_streamlit_style = """
 <style>#root > div:nth-child(1) > div > div > div > div > section > div {padding-top: 0rem;}</style>
@@ -41,20 +43,32 @@ hide_streamlit_style = """
 st.markdown(hide_streamlit_style, unsafe_allow_html=True)
 st.title(f"MoneyPrinterTurbo v{config.project_version}")
 
+support_locales = [
+    "zh-CN",
+    "zh-HK",
+    "zh-TW",
+    "de-DE",
+    "en-US",
+    "fr-FR",
+    "vi-VN",
+    "th-TH",
+]
+
 font_dir = os.path.join(root_dir, "resource", "fonts")
 song_dir = os.path.join(root_dir, "resource", "songs")
 i18n_dir = os.path.join(root_dir, "webui", "i18n")
 config_file = os.path.join(root_dir, "webui", ".streamlit", "webui.toml")
 system_locale = utils.get_system_locale()
+# print(f"******** system locale: {system_locale} ********")
 
-if 'video_subject' not in st.session_state:
-    st.session_state['video_subject'] = ''
-if 'video_script' not in st.session_state:
-    st.session_state['video_script'] = ''
-if 'video_terms' not in st.session_state:
-    st.session_state['video_terms'] = ''
-if 'ui_language' not in st.session_state:
-    st.session_state['ui_language'] = config.ui.get("language", system_locale)
+if "video_subject" not in st.session_state:
+    st.session_state["video_subject"] = ""
+if "video_script" not in st.session_state:
+    st.session_state["video_script"] = ""
+if "video_terms" not in st.session_state:
+    st.session_state["video_terms"] = ""
+if "ui_language" not in st.session_state:
+    st.session_state["ui_language"] = config.ui.get("language", system_locale)
 
 
 def get_all_fonts():
@@ -81,25 +95,25 @@ def open_task_folder(task_id):
         sys = platform.system()
         path = os.path.join(root_dir, "storage", "tasks", task_id)
         if os.path.exists(path):
-            if sys == 'Windows':
+            if sys == "Windows":
                 os.system(f"start {path}")
-            if sys == 'Darwin':
+            if sys == "Darwin":
                 os.system(f"open {path}")
     except Exception as e:
         logger.error(e)
 
 
 def scroll_to_bottom():
-    js = f"""
+    js = """
     <script>
         console.log("scroll_to_bottom");
-        function scroll(dummy_var_to_force_repeat_execution){{
+        function scroll(dummy_var_to_force_repeat_execution){
             var sections = parent.document.querySelectorAll('section.main');
             console.log(sections);
-            for(let index = 0; index<sections.length; index++) {{
+            for(let index = 0; index<sections.length; index++) {
                 sections[index].scrollTop = sections[index].scrollHeight;
-            }}
-        }}
+            }
+        }
         scroll(1);
     </script>
     """
@@ -119,12 +133,15 @@ def init_log():
         record["file"].path = f"./{relative_path}"
         # 返回修改后的格式字符串
         # 您可以根据需要调整这里的格式
-        record['message'] = record['message'].replace(root_dir, ".")
+        record["message"] = record["message"].replace(root_dir, ".")
 
-        _format = '<green>{time:%Y-%m-%d %H:%M:%S}</> | ' + \
-                  '<level>{level}</> | ' + \
-                  '"{file.path}:{line}":<blue> {function}</> ' + \
-                  '- <level>{message}</>' + "\n"
+        _format = (
+            "<green>{time:%Y-%m-%d %H:%M:%S}</> | "
+            + "<level>{level}</> | "
+            + '"{file.path}:{line}":<blue> {function}</> '
+            + "- <level>{message}</>"
+            + "\n"
+        )
         return _format
 
     logger.add(
@@ -141,120 +158,316 @@ locales = utils.load_locales(i18n_dir)
 
 
 def tr(key):
-    loc = locales.get(st.session_state['ui_language'], {})
+    loc = locales.get(st.session_state["ui_language"], {})
     return loc.get("Translation", {}).get(key, key)
 
 
 st.write(tr("Get Help"))
 
-with st.expander(tr("Basic Settings"), expanded=False):
-    config_panels = st.columns(3)
-    left_config_panel = config_panels[0]
-    middle_config_panel = config_panels[1]
-    right_config_panel = config_panels[2]
-    with left_config_panel:
-        display_languages = []
-        selected_index = 0
-        for i, code in enumerate(locales.keys()):
-            display_languages.append(f"{code} - {locales[code].get('Language')}")
-            if code == st.session_state['ui_language']:
-                selected_index = i
+llm_provider = config.app.get("llm_provider", "").lower()
 
-        selected_language = st.selectbox(tr("Language"), options=display_languages,
-                                         index=selected_index)
-        if selected_language:
-            code = selected_language.split(" - ")[0].strip()
-            st.session_state['ui_language'] = code
-            config.ui['language'] = code
+if not config.app.get("hide_config", False):
+    with st.expander(tr("Basic Settings"), expanded=False):
+        config_panels = st.columns(3)
+        left_config_panel = config_panels[0]
+        middle_config_panel = config_panels[1]
+        right_config_panel = config_panels[2]
+        with left_config_panel:
+            display_languages = []
+            selected_index = 0
+            for i, code in enumerate(locales.keys()):
+                display_languages.append(f"{code} - {locales[code].get('Language')}")
+                if code == st.session_state["ui_language"]:
+                    selected_index = i
 
-    with middle_config_panel:
-        #   openai
-        #   moonshot (月之暗面)
-        #   oneapi
-        #   g4f
-        #   azure
-        #   qwen (通义千问)
-        #   gemini
-        #   ollama
-        llm_providers = ['OpenAI', 'Moonshot', 'Azure', 'Qwen', 'Gemini', 'Ollama', 'G4f', 'OneAPI', "Cloudflare"]
-        saved_llm_provider = config.app.get("llm_provider", "OpenAI").lower()
-        saved_llm_provider_index = 0
-        for i, provider in enumerate(llm_providers):
-            if provider.lower() == saved_llm_provider:
-                saved_llm_provider_index = i
-                break
+            selected_language = st.selectbox(
+                tr("Language"), options=display_languages, index=selected_index
+            )
+            if selected_language:
+                code = selected_language.split(" - ")[0].strip()
+                st.session_state["ui_language"] = code
+                config.ui["language"] = code
 
-        llm_provider = st.selectbox(tr("LLM Provider"), options=llm_providers, index=saved_llm_provider_index)
-        llm_provider = llm_provider.lower()
-        config.app["llm_provider"] = llm_provider
+            # 是否禁用日志显示
+            hide_log = st.checkbox(
+                tr("Hide Log"), value=config.app.get("hide_log", False)
+            )
+            config.ui["hide_log"] = hide_log
 
-        llm_api_key = config.app.get(f"{llm_provider}_api_key", "")
-        llm_base_url = config.app.get(f"{llm_provider}_base_url", "")
-        llm_model_name = config.app.get(f"{llm_provider}_model_name", "")
-        llm_account_id = config.app.get(f"{llm_provider}_account_id", "")
-        st_llm_api_key = st.text_input(tr("API Key"), value=llm_api_key, type="password")
-        st_llm_base_url = st.text_input(tr("Base Url"), value=llm_base_url)
-        st_llm_model_name = st.text_input(tr("Model Name"), value=llm_model_name)
-        if st_llm_api_key:
-            config.app[f"{llm_provider}_api_key"] = st_llm_api_key
-        if st_llm_base_url:
-            config.app[f"{llm_provider}_base_url"] = st_llm_base_url
-        if st_llm_model_name:
-            config.app[f"{llm_provider}_model_name"] = st_llm_model_name
+        with middle_config_panel:
+            #   openai
+            #   moonshot (月之暗面)
+            #   oneapi
+            #   g4f
+            #   azure
+            #   qwen (通义千问)
+            #   gemini
+            #   ollama
+            llm_providers = [
+                "OpenAI",
+                "Moonshot",
+                "Azure",
+                "Qwen",
+                "DeepSeek",
+                "Gemini",
+                "Ollama",
+                "G4f",
+                "OneAPI",
+                "Cloudflare",
+                "ERNIE",
+            ]
+            saved_llm_provider = config.app.get("llm_provider", "OpenAI").lower()
+            saved_llm_provider_index = 0
+            for i, provider in enumerate(llm_providers):
+                if provider.lower() == saved_llm_provider:
+                    saved_llm_provider_index = i
+                    break
 
-        if llm_provider == 'cloudflare':
-            st_llm_account_id = st.text_input(tr("Account ID"), value=llm_account_id)
-            if st_llm_account_id:
-                config.app[f"{llm_provider}_account_id"] = st_llm_account_id
+            llm_provider = st.selectbox(
+                tr("LLM Provider"),
+                options=llm_providers,
+                index=saved_llm_provider_index,
+            )
+            llm_helper = st.container()
+            llm_provider = llm_provider.lower()
+            config.app["llm_provider"] = llm_provider
 
-    with right_config_panel:
-        pexels_api_keys = config.app.get("pexels_api_keys", [])
-        if isinstance(pexels_api_keys, str):
-            pexels_api_keys = [pexels_api_keys]
-        pexels_api_key = ", ".join(pexels_api_keys)
+            llm_api_key = config.app.get(f"{llm_provider}_api_key", "")
+            llm_secret_key = config.app.get(
+                f"{llm_provider}_secret_key", ""
+            )  # only for baidu ernie
+            llm_base_url = config.app.get(f"{llm_provider}_base_url", "")
+            llm_model_name = config.app.get(f"{llm_provider}_model_name", "")
+            llm_account_id = config.app.get(f"{llm_provider}_account_id", "")
 
-        pexels_api_key = st.text_input(tr("Pexels API Key"), value=pexels_api_key, type="password")
-        pexels_api_key = pexels_api_key.replace(" ", "")
-        if pexels_api_key:
-            config.app["pexels_api_keys"] = pexels_api_key.split(",")
+            tips = ""
+            if llm_provider == "ollama":
+                if not llm_model_name:
+                    llm_model_name = "qwen:7b"
+                if not llm_base_url:
+                    llm_base_url = "http://localhost:11434/v1"
+
+                with llm_helper:
+                    tips = """
+                           ##### Ollama配置说明
+                           - **API Key**: 随便填写，比如 123
+                           - **Base Url**: 一般为 http://localhost:11434/v1
+                              - 如果 `MoneyPrinterTurbo` 和 `Ollama` **不在同一台机器上**，需要填写 `Ollama` 机器的IP地址
+                              - 如果 `MoneyPrinterTurbo` 是 `Docker` 部署，建议填写 `http://host.docker.internal:11434/v1`
+                           - **Model Name**: 使用 `ollama list` 查看，比如 `qwen:7b`
+                           """
+
+            if llm_provider == "openai":
+                if not llm_model_name:
+                    llm_model_name = "gpt-3.5-turbo"
+                with llm_helper:
+                    tips = """
+                           ##### OpenAI 配置说明
+                           > 需要VPN开启全局流量模式
+                           - **API Key**: [点击到官网申请](https://platform.openai.com/api-keys)
+                           - **Base Url**: 可以留空
+                           - **Model Name**: 填写**有权限**的模型，[点击查看模型列表](https://platform.openai.com/settings/organization/limits)
+                           """
+
+            if llm_provider == "moonshot":
+                if not llm_model_name:
+                    llm_model_name = "moonshot-v1-8k"
+                with llm_helper:
+                    tips = """
+                           ##### Moonshot 配置说明
+                           - **API Key**: [点击到官网申请](https://platform.moonshot.cn/console/api-keys)
+                           - **Base Url**: 固定为 https://api.moonshot.cn/v1
+                           - **Model Name**: 比如 moonshot-v1-8k，[点击查看模型列表](https://platform.moonshot.cn/docs/intro#%E6%A8%A1%E5%9E%8B%E5%88%97%E8%A1%A8)
+                           """
+            if llm_provider == "oneapi":
+                if not llm_model_name:
+                    llm_model_name = (
+                        "claude-3-5-sonnet-20240620"  # 默认模型，可以根据需要调整
+                    )
+                with llm_helper:
+                    tips = """
+                        ##### OneAPI 配置说明
+                        - **API Key**: 填写您的 OneAPI 密钥
+                        - **Base Url**: 填写 OneAPI 的基础 URL
+                        - **Model Name**: 填写您要使用的模型名称，例如 claude-3-5-sonnet-20240620
+                        """
+
+            if llm_provider == "qwen":
+                if not llm_model_name:
+                    llm_model_name = "qwen-max"
+                with llm_helper:
+                    tips = """
+                           ##### 通义千问Qwen 配置说明
+                           - **API Key**: [点击到官网申请](https://dashscope.console.aliyun.com/apiKey)
+                           - **Base Url**: 留空
+                           - **Model Name**: 比如 qwen-max，[点击查看模型列表](https://help.aliyun.com/zh/dashscope/developer-reference/model-introduction#3ef6d0bcf91wy)
+                           """
+
+            if llm_provider == "g4f":
+                if not llm_model_name:
+                    llm_model_name = "gpt-3.5-turbo"
+                with llm_helper:
+                    tips = """
+                           ##### gpt4free 配置说明
+                           > [GitHub开源项目](https://github.com/xtekky/gpt4free)，可以免费使用GPT模型，但是**稳定性较差**
+                           - **API Key**: 随便填写，比如 123
+                           - **Base Url**: 留空
+                           - **Model Name**: 比如 gpt-3.5-turbo，[点击查看模型列表](https://github.com/xtekky/gpt4free/blob/main/g4f/models.py#L308)
+                           """
+            if llm_provider == "azure":
+                with llm_helper:
+                    tips = """
+                           ##### Azure 配置说明
+                           > [点击查看如何部署模型](https://learn.microsoft.com/zh-cn/azure/ai-services/openai/how-to/create-resource)
+                           - **API Key**: [点击到Azure后台创建](https://portal.azure.com/#view/Microsoft_Azure_ProjectOxford/CognitiveServicesHub/~/OpenAI)
+                           - **Base Url**: 留空
+                           - **Model Name**: 填写你实际的部署名
+                           """
+
+            if llm_provider == "gemini":
+                if not llm_model_name:
+                    llm_model_name = "gemini-1.0-pro"
+
+                with llm_helper:
+                    tips = """
+                            ##### Gemini 配置说明
+                            > 需要VPN开启全局流量模式
+                           - **API Key**: [点击到官网申请](https://ai.google.dev/)
+                           - **Base Url**: 留空
+                           - **Model Name**: 比如 gemini-1.0-pro
+                           """
+
+            if llm_provider == "deepseek":
+                if not llm_model_name:
+                    llm_model_name = "deepseek-chat"
+                if not llm_base_url:
+                    llm_base_url = "https://api.deepseek.com"
+                with llm_helper:
+                    tips = """
+                           ##### DeepSeek 配置说明
+                           - **API Key**: [点击到官网申请](https://platform.deepseek.com/api_keys)
+                           - **Base Url**: 固定为 https://api.deepseek.com
+                           - **Model Name**: 固定为 deepseek-chat
+                           """
+
+            if llm_provider == "ernie":
+                with llm_helper:
+                    tips = """
+                           ##### 百度文心一言 配置说明
+                           - **API Key**: [点击到官网申请](https://console.bce.baidu.com/qianfan/ais/console/applicationConsole/application)
+                           - **Secret Key**: [点击到官网申请](https://console.bce.baidu.com/qianfan/ais/console/applicationConsole/application)
+                           - **Base Url**: 填写 **请求地址** [点击查看文档](https://cloud.baidu.com/doc/WENXINWORKSHOP/s/jlil56u11#%E8%AF%B7%E6%B1%82%E8%AF%B4%E6%98%8E)
+                           """
+
+            if tips and config.ui["language"] == "zh":
+                st.warning(
+                    "中国用户建议使用 **DeepSeek** 或 **Moonshot** 作为大模型提供商\n- 国内可直接访问，不需要VPN \n- 注册就送额度，基本够用"
+                )
+                st.info(tips)
+
+            st_llm_api_key = st.text_input(
+                tr("API Key"), value=llm_api_key, type="password"
+            )
+            st_llm_base_url = st.text_input(tr("Base Url"), value=llm_base_url)
+            st_llm_model_name = ""
+            if llm_provider != "ernie":
+                st_llm_model_name = st.text_input(
+                    tr("Model Name"),
+                    value=llm_model_name,
+                    key=f"{llm_provider}_model_name_input",
+                )
+                if st_llm_model_name:
+                    config.app[f"{llm_provider}_model_name"] = st_llm_model_name
+            else:
+                st_llm_model_name = None
+
+            if st_llm_api_key:
+                config.app[f"{llm_provider}_api_key"] = st_llm_api_key
+            if st_llm_base_url:
+                config.app[f"{llm_provider}_base_url"] = st_llm_base_url
+            if st_llm_model_name:
+                config.app[f"{llm_provider}_model_name"] = st_llm_model_name
+            if llm_provider == "ernie":
+                st_llm_secret_key = st.text_input(
+                    tr("Secret Key"), value=llm_secret_key, type="password"
+                )
+                config.app[f"{llm_provider}_secret_key"] = st_llm_secret_key
+
+            if llm_provider == "cloudflare":
+                st_llm_account_id = st.text_input(
+                    tr("Account ID"), value=llm_account_id
+                )
+                if st_llm_account_id:
+                    config.app[f"{llm_provider}_account_id"] = st_llm_account_id
+
+        with right_config_panel:
+
+            def get_keys_from_config(cfg_key):
+                api_keys = config.app.get(cfg_key, [])
+                if isinstance(api_keys, str):
+                    api_keys = [api_keys]
+                api_key = ", ".join(api_keys)
+                return api_key
+
+            def save_keys_to_config(cfg_key, value):
+                value = value.replace(" ", "")
+                if value:
+                    config.app[cfg_key] = value.split(",")
+
+            pexels_api_key = get_keys_from_config("pexels_api_keys")
+            pexels_api_key = st.text_input(
+                tr("Pexels API Key"), value=pexels_api_key, type="password"
+            )
+            save_keys_to_config("pexels_api_keys", pexels_api_key)
+
+            pixabay_api_key = get_keys_from_config("pixabay_api_keys")
+            pixabay_api_key = st.text_input(
+                tr("Pixabay API Key"), value=pixabay_api_key, type="password"
+            )
+            save_keys_to_config("pixabay_api_keys", pixabay_api_key)
 
 panel = st.columns(3)
 left_panel = panel[0]
 middle_panel = panel[1]
 right_panel = panel[2]
 
-params = VideoParams()
+params = VideoParams(video_subject="")
+uploaded_files = []
 
 with left_panel:
     with st.container(border=True):
         st.write(tr("Video Script Settings"))
-        params.video_subject = st.text_input(tr("Video Subject"),
-                                             value=st.session_state['video_subject']).strip()
+        params.video_subject = st.text_input(
+            tr("Video Subject"), value=st.session_state["video_subject"]
+        ).strip()
 
         video_languages = [
             (tr("Auto Detect"), ""),
         ]
-        for code in ["zh-CN", "zh-TW", "de-DE", "en-US"]:
+        for code in support_locales:
             video_languages.append((code, code))
 
-        selected_index = st.selectbox(tr("Script Language"),
-                                      index=0,
-                                      options=range(len(video_languages)),  # 使用索引作为内部选项值
-                                      format_func=lambda x: video_languages[x][0]  # 显示给用户的是标签
-                                      )
+        selected_index = st.selectbox(
+            tr("Script Language"),
+            index=0,
+            options=range(len(video_languages)),  # 使用索引作为内部选项值
+            format_func=lambda x: video_languages[x][0],  # 显示给用户的是标签
+        )
         params.video_language = video_languages[selected_index][1]
 
-        if st.button(tr("Generate Video Script and Keywords"), key="auto_generate_script"):
+        if st.button(
+            tr("Generate Video Script and Keywords"), key="auto_generate_script"
+        ):
             with st.spinner(tr("Generating Video Script and Keywords")):
-                script = llm.generate_script(video_subject=params.video_subject, language=params.video_language)
+                script = llm.generate_script(
+                    video_subject=params.video_subject, language=params.video_language
+                )
                 terms = llm.generate_terms(params.video_subject, script)
-                st.session_state['video_script'] = script
-                st.session_state['video_terms'] = ", ".join(terms)
+                st.session_state["video_script"] = script
+                st.session_state["video_terms"] = ", ".join(terms)
 
         params.video_script = st.text_area(
-            tr("Video Script"),
-            value=st.session_state['video_script'],
-            height=280
+            tr("Video Script"), value=st.session_state["video_script"], height=280
         )
         if st.button(tr("Generate Video Keywords"), key="auto_generate_terms"):
             if not params.video_script:
@@ -263,12 +476,11 @@ with left_panel:
 
             with st.spinner(tr("Generating Video Keywords")):
                 terms = llm.generate_terms(params.video_subject, params.video_script)
-                st.session_state['video_terms'] = ", ".join(terms)
+                st.session_state["video_terms"] = ", ".join(terms)
 
         params.video_terms = st.text_area(
-            tr("Video Keywords"),
-            value=st.session_state['video_terms'],
-            height=50)
+            tr("Video Keywords"), value=st.session_state["video_terms"], height=50
+        )
 
 with middle_panel:
     with st.container(border=True):
@@ -277,83 +489,183 @@ with middle_panel:
             (tr("Sequential"), "sequential"),
             (tr("Random"), "random"),
         ]
-        selected_index = st.selectbox(tr("Video Concat Mode"),
-                                      index=1,
-                                      options=range(len(video_concat_modes)),  # 使用索引作为内部选项值
-                                      format_func=lambda x: video_concat_modes[x][0]  # 显示给用户的是标签
-                                      )
-        params.video_concat_mode = VideoConcatMode(video_concat_modes[selected_index][1])
+        video_sources = [
+            (tr("Pexels"), "pexels"),
+            (tr("Pixabay"), "pixabay"),
+            (tr("Local file"), "local"),
+            (tr("TikTok"), "douyin"),
+            (tr("Bilibili"), "bilibili"),
+            (tr("Xiaohongshu"), "xiaohongshu"),
+        ]
+
+        saved_video_source_name = config.app.get("video_source", "pexels")
+        saved_video_source_index = [v[1] for v in video_sources].index(
+            saved_video_source_name
+        )
+
+        selected_index = st.selectbox(
+            tr("Video Source"),
+            options=range(len(video_sources)),
+            format_func=lambda x: video_sources[x][0],
+            index=saved_video_source_index,
+        )
+        params.video_source = video_sources[selected_index][1]
+        config.app["video_source"] = params.video_source
+
+        if params.video_source == "local":
+            _supported_types = FILE_TYPE_VIDEOS + FILE_TYPE_IMAGES
+            uploaded_files = st.file_uploader(
+                "Upload Local Files",
+                type=["mp4", "mov", "avi", "flv", "mkv", "jpg", "jpeg", "png"],
+                accept_multiple_files=True,
+            )
+
+        selected_index = st.selectbox(
+            tr("Video Concat Mode"),
+            index=1,
+            options=range(len(video_concat_modes)),  # 使用索引作为内部选项值
+            format_func=lambda x: video_concat_modes[x][0],  # 显示给用户的是标签
+        )
+        params.video_concat_mode = VideoConcatMode(
+            video_concat_modes[selected_index][1]
+        )
 
         video_aspect_ratios = [
             (tr("Portrait"), VideoAspect.portrait.value),
             (tr("Landscape"), VideoAspect.landscape.value),
         ]
-        selected_index = st.selectbox(tr("Video Ratio"),
-                                      options=range(len(video_aspect_ratios)),  # 使用索引作为内部选项值
-                                      format_func=lambda x: video_aspect_ratios[x][0]  # 显示给用户的是标签
-                                      )
+        selected_index = st.selectbox(
+            tr("Video Ratio"),
+            options=range(len(video_aspect_ratios)),  # 使用索引作为内部选项值
+            format_func=lambda x: video_aspect_ratios[x][0],  # 显示给用户的是标签
+        )
         params.video_aspect = VideoAspect(video_aspect_ratios[selected_index][1])
 
-        params.video_clip_duration = st.selectbox(tr("Clip Duration"), options=[2, 3, 4, 5, 6], index=1)
-        params.video_count = st.selectbox(tr("Number of Videos Generated Simultaneously"), options=[1, 2, 3, 4, 5],
-                                          index=0)
+        params.video_clip_duration = st.selectbox(
+            tr("Clip Duration"), options=[2, 3, 4, 5, 6, 7, 8, 9, 10], index=1
+        )
+        params.video_count = st.selectbox(
+            tr("Number of Videos Generated Simultaneously"),
+            options=[1, 2, 3, 4, 5],
+            index=0,
+        )
     with st.container(border=True):
         st.write(tr("Audio Settings"))
-        voices = voice.get_all_azure_voices(filter_locals=["zh-CN", "zh-HK", "zh-TW", "de-DE", "en-US", "fr-FR"])
+
+        # tts_providers = ['edge', 'azure']
+        # tts_provider = st.selectbox(tr("TTS Provider"), tts_providers)
+
+        voices = voice.get_all_azure_voices(filter_locals=support_locales)
         friendly_names = {
-            v: v.
-            replace("Female", tr("Female")).
-            replace("Male", tr("Male")).
-            replace("Neural", "") for
-            v in voices}
+            v: v.replace("Female", tr("Female"))
+            .replace("Male", tr("Male"))
+            .replace("Neural", "")
+            for v in voices
+        }
         saved_voice_name = config.ui.get("voice_name", "")
         saved_voice_name_index = 0
         if saved_voice_name in friendly_names:
             saved_voice_name_index = list(friendly_names.keys()).index(saved_voice_name)
         else:
             for i, v in enumerate(voices):
-                if v.lower().startswith(st.session_state['ui_language'].lower()):
+                if (
+                    v.lower().startswith(st.session_state["ui_language"].lower())
+                    and "V2" not in v
+                ):
                     saved_voice_name_index = i
                     break
 
-        selected_friendly_name = st.selectbox(tr("Speech Synthesis"),
-                                              options=list(friendly_names.values()),
-                                              index=saved_voice_name_index)
+        selected_friendly_name = st.selectbox(
+            tr("Speech Synthesis"),
+            options=list(friendly_names.values()),
+            index=saved_voice_name_index,
+        )
 
-        voice_name = list(friendly_names.keys())[list(friendly_names.values()).index(selected_friendly_name)]
+        voice_name = list(friendly_names.keys())[
+            list(friendly_names.values()).index(selected_friendly_name)
+        ]
         params.voice_name = voice_name
-        config.ui['voice_name'] = voice_name
+        config.ui["voice_name"] = voice_name
+
+        if st.button(tr("Play Voice")):
+            play_content = params.video_subject
+            if not play_content:
+                play_content = params.video_script
+            if not play_content:
+                play_content = tr("Voice Example")
+            with st.spinner(tr("Synthesizing Voice")):
+                temp_dir = utils.storage_dir("temp", create=True)
+                audio_file = os.path.join(temp_dir, f"tmp-voice-{str(uuid4())}.mp3")
+                sub_maker = voice.tts(
+                    text=play_content,
+                    voice_name=voice_name,
+                    voice_rate=params.voice_rate,
+                    voice_file=audio_file,
+                )
+                # if the voice file generation failed, try again with a default content.
+                if not sub_maker:
+                    play_content = "This is a example voice. if you hear this, the voice synthesis failed with the original content."
+                    sub_maker = voice.tts(
+                        text=play_content,
+                        voice_name=voice_name,
+                        voice_rate=params.voice_rate,
+                        voice_file=audio_file,
+                    )
+
+                if sub_maker and os.path.exists(audio_file):
+                    st.audio(audio_file, format="audio/mp3")
+                    if os.path.exists(audio_file):
+                        os.remove(audio_file)
+
         if voice.is_azure_v2_voice(voice_name):
-            saved_azure_speech_region = config.azure.get(f"speech_region", "")
-            saved_azure_speech_key = config.azure.get(f"speech_key", "")
-            azure_speech_region = st.text_input(tr("Speech Region"), value=saved_azure_speech_region)
-            azure_speech_key = st.text_input(tr("Speech Key"), value=saved_azure_speech_key, type="password")
+            saved_azure_speech_region = config.azure.get("speech_region", "")
+            saved_azure_speech_key = config.azure.get("speech_key", "")
+            azure_speech_region = st.text_input(
+                tr("Speech Region"), value=saved_azure_speech_region
+            )
+            azure_speech_key = st.text_input(
+                tr("Speech Key"), value=saved_azure_speech_key, type="password"
+            )
             config.azure["speech_region"] = azure_speech_region
             config.azure["speech_key"] = azure_speech_key
 
-        params.voice_volume = st.selectbox(tr("Speech Volume"),
-                                           options=[0.6, 0.8, 1.0, 1.2, 1.5, 2.0, 3.0, 4.0, 5.0], index=2)
+        params.voice_volume = st.selectbox(
+            tr("Speech Volume"),
+            options=[0.6, 0.8, 1.0, 1.2, 1.5, 2.0, 3.0, 4.0, 5.0],
+            index=2,
+        )
+
+        params.voice_rate = st.selectbox(
+            tr("Speech Rate"),
+            options=[0.8, 0.9, 1.0, 1.1, 1.2, 1.3, 1.5, 1.8, 2.0],
+            index=2,
+        )
+
         bgm_options = [
             (tr("No Background Music"), ""),
             (tr("Random Background Music"), "random"),
             (tr("Custom Background Music"), "custom"),
         ]
-        selected_index = st.selectbox(tr("Background Music"),
-                                      index=1,
-                                      options=range(len(bgm_options)),  # 使用索引作为内部选项值
-                                      format_func=lambda x: bgm_options[x][0]  # 显示给用户的是标签
-                                      )
+        selected_index = st.selectbox(
+            tr("Background Music"),
+            index=1,
+            options=range(len(bgm_options)),  # 使用索引作为内部选项值
+            format_func=lambda x: bgm_options[x][0],  # 显示给用户的是标签
+        )
         # 获取选择的背景音乐类型
-        bgm_type = bgm_options[selected_index][1]
+        params.bgm_type = bgm_options[selected_index][1]
 
         # 根据选择显示或隐藏组件
-        if bgm_type == "custom":
+        if params.bgm_type == "custom":
             custom_bgm_file = st.text_input(tr("Custom Background Music File"))
             if custom_bgm_file and os.path.exists(custom_bgm_file):
                 params.bgm_file = custom_bgm_file
                 # st.write(f":red[已选择自定义背景音乐]：**{custom_bgm_file}**")
-        params.bgm_volume = st.selectbox(tr("Background Music Volume"),
-                                         options=[0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0], index=2)
+        params.bgm_volume = st.selectbox(
+            tr("Background Music Volume"),
+            options=[0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0],
+            index=2,
+        )
 
 with right_panel:
     with st.container(border=True):
@@ -364,31 +676,48 @@ with right_panel:
         saved_font_name_index = 0
         if saved_font_name in font_names:
             saved_font_name_index = font_names.index(saved_font_name)
-        params.font_name = st.selectbox(tr("Font"), font_names, index=saved_font_name_index)
-        config.ui['font_name'] = params.font_name
+        params.font_name = st.selectbox(
+            tr("Font"), font_names, index=saved_font_name_index
+        )
+        config.ui["font_name"] = params.font_name
 
         subtitle_positions = [
             (tr("Top"), "top"),
             (tr("Center"), "center"),
             (tr("Bottom"), "bottom"),
+            (tr("Custom"), "custom"),
         ]
-        selected_index = st.selectbox(tr("Position"),
-                                      index=2,
-                                      options=range(len(subtitle_positions)),  # 使用索引作为内部选项值
-                                      format_func=lambda x: subtitle_positions[x][0]  # 显示给用户的是标签
-                                      )
+        selected_index = st.selectbox(
+            tr("Position"),
+            index=2,
+            options=range(len(subtitle_positions)),
+            format_func=lambda x: subtitle_positions[x][0],
+        )
         params.subtitle_position = subtitle_positions[selected_index][1]
 
+        if params.subtitle_position == "custom":
+            custom_position = st.text_input(
+                tr("Custom Position (% from top)"), value="70.0"
+            )
+            try:
+                params.custom_position = float(custom_position)
+                if params.custom_position < 0 or params.custom_position > 100:
+                    st.error(tr("Please enter a value between 0 and 100"))
+            except ValueError:
+                st.error(tr("Please enter a valid number"))
+
         font_cols = st.columns([0.3, 0.7])
         with font_cols[0]:
             saved_text_fore_color = config.ui.get("text_fore_color", "#FFFFFF")
-            params.text_fore_color = st.color_picker(tr("Font Color"), saved_text_fore_color)
-            config.ui['text_fore_color'] = params.text_fore_color
+            params.text_fore_color = st.color_picker(
+                tr("Font Color"), saved_text_fore_color
+            )
+            config.ui["text_fore_color"] = params.text_fore_color
 
         with font_cols[1]:
             saved_font_size = config.ui.get("font_size", 60)
             params.font_size = st.slider(tr("Font Size"), 30, 100, saved_font_size)
-            config.ui['font_size'] = params.font_size
+            config.ui["font_size"] = params.font_size
 
         stroke_cols = st.columns([0.3, 0.7])
         with stroke_cols[0]:
@@ -405,26 +734,49 @@ if start_button:
         scroll_to_bottom()
         st.stop()
 
-    if llm_provider != 'g4f' and not config.app.get(f"{llm_provider}_api_key", ""):
+    if llm_provider != "g4f" and not config.app.get(f"{llm_provider}_api_key", ""):
         st.error(tr("Please Enter the LLM API Key"))
         scroll_to_bottom()
         st.stop()
 
-    if not config.app.get("pexels_api_keys", ""):
+    if params.video_source not in ["pexels", "pixabay", "local"]:
+        st.error(tr("Please Select a Valid Video Source"))
+        scroll_to_bottom()
+        st.stop()
+
+    if params.video_source == "pexels" and not config.app.get("pexels_api_keys", ""):
         st.error(tr("Please Enter the Pexels API Key"))
         scroll_to_bottom()
         st.stop()
 
+    if params.video_source == "pixabay" and not config.app.get("pixabay_api_keys", ""):
+        st.error(tr("Please Enter the Pixabay API Key"))
+        scroll_to_bottom()
+        st.stop()
+
+    if uploaded_files:
+        local_videos_dir = utils.storage_dir("local_videos", create=True)
+        for file in uploaded_files:
+            file_path = os.path.join(local_videos_dir, f"{file.file_id}_{file.name}")
+            with open(file_path, "wb") as f:
+                f.write(file.getbuffer())
+                m = MaterialInfo()
+                m.provider = "local"
+                m.url = file_path
+                if not params.video_materials:
+                    params.video_materials = []
+                params.video_materials.append(m)
+
     log_container = st.empty()
     log_records = []
 
-
     def log_received(msg):
+        if config.ui["hide_log"]:
+            return
         with log_container:
             log_records.append(msg)
             st.code("\n".join(log_records))
 
-
     logger.add(log_received)
 
     st.toast(tr("Generating Video"))
@@ -433,16 +785,20 @@ if start_button:
     scroll_to_bottom()
 
     result = tm.start(task_id=task_id, params=params)
+    if not result or "videos" not in result:
+        st.error(tr("Video Generation Failed"))
+        logger.error(tr("Video Generation Failed"))
+        scroll_to_bottom()
+        st.stop()
 
     video_files = result.get("videos", [])
     st.success(tr("Video Generation Completed"))
     try:
         if video_files:
-            # center the video player
             player_cols = st.columns(len(video_files) * 2 + 1)
             for i, url in enumerate(video_files):
                 player_cols[i * 2 + 1].video(url)
-    except Exception as e:
+    except Exception:
         pass
 
     open_task_folder(task_id)

webui/i18n/de.json

@@ -26,6 +26,7 @@
     "Speech Region": "Region(:red[Required，[Get Region](https://portal.azure.com/#view/Microsoft_Azure_ProjectOxford/CognitiveServicesHub/~/SpeechServices)])",
     "Speech Key": "API Key(:red[Required，[Get API Key](https://portal.azure.com/#view/Microsoft_Azure_ProjectOxford/CognitiveServicesHub/~/SpeechServices)])",
     "Speech Volume": "Lautstärke der Sprachausgabe",
+    "Speech Rate": "Lesegeschwindigkeit (1,0 bedeutet 1x)",
     "Male": "Männlich",
     "Female": "Weiblich",
     "Background Music": "Hintergrundmusik",
@@ -41,6 +42,7 @@
     "Top": "Oben",
     "Center": "Mittig",
     "Bottom": "Unten (empfohlen)",
+    "Custom": "Benutzerdefinierte Position (70, was 70% von oben bedeutet)",
     "Font Size": "Schriftgröße für Untertitel",
     "Font Color": "Schriftfarbe",
     "Stroke Color": "Kontur",
@@ -50,9 +52,11 @@
     "Generating Video": "Video wird erstellt, bitte warten...",
     "Start Generating Video": "Beginne mit der Generierung",
     "Video Generation Completed": "Video erfolgreich generiert",
+    "Video Generation Failed": "Video Generierung fehlgeschlagen",
     "You can download the generated video from the following links": "Sie können das generierte Video über die folgenden Links herunterladen",
     "Basic Settings": "**Grunde Instellungen**",
-    "Pexels API Key": "Pexels API Key (:red[Required] [Get API Key](https://www.pexels.com/api/))",
+    "Pexels API Key": "Pexels API Key ([Get API Key](https://www.pexels.com/api/))",
+    "Pixabay API Key": "Pixabay API Key ([Get API Key](https://pixabay.com/api/docs/#api_search_videos))",
     "Language": "Language",
     "LLM Provider": "LLM Provider",
     "API Key": "API Key (:red[Required])",
@@ -60,6 +64,16 @@
     "Model Name": "Model Name",
     "Please Enter the LLM API Key": "Please Enter the **LLM API Key**",
     "Please Enter the Pexels API Key": "Please Enter the **Pexels API Key**",
-    "Get Help": "If you need help, or have any questions, you can join discord for help: https://harryai.cc"
+    "Please Enter the Pixabay API Key": "Please Enter the **Pixabay API Key**",
+    "Get Help": "If you need help, or have any questions, you can join discord for help: https://harryai.cc",
+    "Video Source": "Video Source",
+    "TikTok": "TikTok (TikTok support is coming soon)",
+    "Bilibili": "Bilibili (Bilibili support is coming soon)",
+    "Xiaohongshu": "Xiaohongshu (Xiaohongshu support is coming soon)",
+    "Local file": "Local file",
+    "Play Voice": "Play Voice",
+    "Voice Example": "This is an example text for testing speech synthesis",
+    "Synthesizing Voice": "Synthesizing voice, please wait...",
+    "TTS Provider": "Select the voice synthesis provider"
   }
 }

webui/i18n/en.json

@@ -26,6 +26,7 @@
     "Speech Region": "Region(:red[Required，[Get Region](https://portal.azure.com/#view/Microsoft_Azure_ProjectOxford/CognitiveServicesHub/~/SpeechServices)])",
     "Speech Key": "API Key(:red[Required，[Get API Key](https://portal.azure.com/#view/Microsoft_Azure_ProjectOxford/CognitiveServicesHub/~/SpeechServices)])",
     "Speech Volume": "Speech Volume (1.0 represents 100%)",
+    "Speech Rate": "Speech Rate (1.0 means 1x speed)",
     "Male": "Male",
     "Female": "Female",
     "Background Music": "Background Music",
@@ -41,6 +42,7 @@
     "Top": "Top",
     "Center": "Center",
     "Bottom": "Bottom (Recommended)",
+    "Custom": "Custom position (70, indicating 70% down from the top)",
     "Font Size": "Subtitle Font Size",
     "Font Color": "Subtitle Font Color",
     "Stroke Color": "Subtitle Outline Color",
@@ -50,8 +52,10 @@
     "Generating Video": "Generating video, please wait...",
     "Start Generating Video": "Start Generating Video",
     "Video Generation Completed": "Video Generation Completed",
+    "Video Generation Failed": "Video Generation Failed",
     "You can download the generated video from the following links": "You can download the generated video from the following links",
-    "Pexels API Key": "Pexels API Key (:red[Required] [Get API Key](https://www.pexels.com/api/))",
+    "Pexels API Key": "Pexels API Key ([Get API Key](https://www.pexels.com/api/))",
+    "Pixabay API Key": "Pixabay API Key ([Get API Key](https://pixabay.com/api/docs/#api_search_videos))",
     "Basic Settings": "**Basic Settings** (:blue[Click to expand])",
     "Language": "Language",
     "LLM Provider": "LLM Provider",
@@ -61,6 +65,17 @@
     "Model Name": "Model Name",
     "Please Enter the LLM API Key": "Please Enter the **LLM API Key**",
     "Please Enter the Pexels API Key": "Please Enter the **Pexels API Key**",
-    "Get Help": "If you need help, or have any questions, you can join discord for help: https://harryai.cc"
+    "Please Enter the Pixabay API Key": "Please Enter the **Pixabay API Key**",
+    "Get Help": "If you need help, or have any questions, you can join discord for help: https://harryai.cc",
+    "Video Source": "Video Source",
+    "TikTok": "TikTok (TikTok support is coming soon)",
+    "Bilibili": "Bilibili (Bilibili support is coming soon)",
+    "Xiaohongshu": "Xiaohongshu (Xiaohongshu support is coming soon)",
+    "Local file": "Local file",
+    "Play Voice": "Play Voice",
+    "Voice Example": "This is an example text for testing speech synthesis",
+    "Synthesizing Voice": "Synthesizing voice, please wait...",
+    "TTS Provider": "Select the voice synthesis provider",
+    "Hide Log": "Hide Log"
   }
 }

webui/i18n/vi.json

@@ -0,0 +1,80 @@
+{
+  "Language": "Tiếng Việt",
+  "Translation": {
+    "Video Script Settings": "**Cài Đặt Kịch Bản Video**",
+    "Video Subject": "Chủ Đề Video (Cung cấp một từ khóa, :red[AI sẽ tự động tạo ra] kịch bản video)",
+    "Script Language": "Ngôn Ngữ cho Việc Tạo Kịch Bản Video (AI sẽ tự động xuất ra dựa trên ngôn ngữ của chủ đề của bạn)",
+    "Generate Video Script and Keywords": "Nhấn để sử dụng AI để tạo [Kịch Bản Video] và [Từ Khóa Video] dựa trên **chủ đề**",
+    "Auto Detect": "Tự Động Phát Hiện",
+    "Video Script": "Kịch Bản Video (:blue[① Tùy chọn, AI tạo ra  ② Dấu câu chính xác giúp việc tạo phụ đề)",
+    "Generate Video Keywords": "Nhấn để sử dụng AI để tạo [Từ Khóa Video] dựa trên **kịch bản**",
+    "Please Enter the Video Subject": "Vui lòng Nhập Kịch Bản Video Trước",
+    "Generating Video Script and Keywords": "AI đang tạo kịch bản video và từ khóa...",
+    "Generating Video Keywords": "AI đang tạo từ khóa video...",
+    "Video Keywords": "Từ Khóa Video (:blue[① Tùy chọn, AI tạo ra ② Sử dụng dấu phẩy **Tiếng Anh** để phân tách, chỉ sử dụng Tiếng Anh])",
+    "Video Settings": "**Cài Đặt Video**",
+    "Video Concat Mode": "Chế Độ Nối Video",
+    "Random": "Nối Ngẫu Nhiên (Được Khuyến Nghị)",
+    "Sequential": "Nối Theo Thứ Tự",
+    "Video Ratio": "Tỷ Lệ Khung Hình Video",
+    "Portrait": "Dọc 9:16",
+    "Landscape": "Ngang 16:9",
+    "Clip Duration": "Thời Lượng Tối Đa Của Đoạn Video (giây)",
+    "Number of Videos Generated Simultaneously": "Số Video Được Tạo Ra Đồng Thời",
+    "Audio Settings": "**Cài Đặt Âm Thanh**",
+    "Speech Synthesis": "Giọng Đọc Văn Bản",
+    "Speech Region": "Vùng(:red[Bắt Buộc，[Lấy Vùng](https://portal.azure.com/#view/Microsoft_Azure_ProjectOxford/CognitiveServicesHub/~/SpeechServices)])",
+    "Speech Key": "Khóa API(:red[Bắt Buộc，[Lấy Khóa API](https://portal.azure.com/#view/Microsoft_Azure_ProjectOxford/CognitiveServicesHub/~/SpeechServices)])",
+    "Speech Volume": "Âm Lượng Giọng Đọc (1.0 đại diện cho 100%)",
+    "Speech Rate": "Tốc độ đọc (1.0 biểu thị tốc độ gốc)",
+    "Male": "Nam",
+    "Female": "Nữ",
+    "Background Music": "Âm Nhạc Nền",
+    "No Background Music": "Không Có Âm Nhạc Nền",
+    "Random Background Music": "Âm Nhạc Nền Ngẫu Nhiên",
+    "Custom Background Music": "Âm Nhạc Nền Tùy Chỉnh",
+    "Custom Background Music File": "Vui lòng nhập đường dẫn tệp cho âm nhạc nền tùy chỉnh:",
+    "Background Music Volume": "Âm Lượng Âm Nhạc Nền (0.2 đại diện cho 20%, âm nhạc nền không nên quá to)",
+    "Subtitle Settings": "**Cài Đặt Phụ Đề**",
+    "Enable Subtitles": "Bật Phụ Đề (Nếu không chọn, các cài đặt dưới đây sẽ không có hiệu lực)",
+    "Font": "Phông Chữ Phụ Đề",
+    "Position": "Vị Trí Phụ Đề",
+    "Top": "Trên",
+    "Center": "Giữa",
+    "Bottom": "Dưới (Được Khuyến Nghị)",
+    "Custom": "Vị trí tùy chỉnh (70, chỉ ra là cách đầu trang 70%)",
+    "Font Size": "Cỡ Chữ Phụ Đề",
+    "Font Color": "Màu Chữ Phụ Đề",
+    "Stroke Color": "Màu Viền Phụ Đề",
+    "Stroke Width": "Độ Rộng Viền Phụ Đề",
+    "Generate Video": "Tạo Video",
+    "Video Script and Subject Cannot Both Be Empty": "Chủ Đề Video và Kịch Bản Video không thể cùng trống",
+    "Generating Video": "Đang tạo video, vui lòng đợi...",
+    "Start Generating Video": "Bắt Đầu Tạo Video",
+    "Video Generation Completed": "Hoàn Tất Tạo Video",
+    "Video Generation Failed": "Tạo Video Thất Bại",
+    "You can download the generated video from the following links": "Bạn có thể tải video được tạo ra từ các liên kết sau",
+    "Pexels API Key": "Khóa API Pexels ([Lấy Khóa API](https://www.pexels.com/api/))",
+    "Pixabay API Key": "Pixabay API Key ([Get API Key](https://pixabay.com/api/docs/#api_search_videos))",
+    "Basic Settings": "**Cài Đặt Cơ Bản** (:blue[Nhấp để mở rộng])",
+    "Language": "Ngôn Ngữ",
+    "LLM Provider": "Nhà Cung Cấp LLM",
+    "API Key": "Khóa API (:red[Bắt Buộc])",
+    "Base Url": "Url Cơ Bản",
+    "Account ID": "ID Tài Khoản (Lấy từ bảng điều khiển Cloudflare)",
+    "Model Name": "Tên Mô Hình",
+    "Please Enter the LLM API Key": "Vui lòng Nhập **Khóa API LLM**",
+    "Please Enter the Pexels API Key": "Vui lòng Nhập **Khóa API Pexels**",
+    "Please Enter the Pixabay API Key": "Vui lòng Nhập **Pixabay API Key**",
+    "Get Help": "Nếu bạn cần giúp đỡ hoặc có bất kỳ câu hỏi nào, bạn có thể tham gia discord để được giúp đỡ: https://harryai.cc",
+    "Video Source": "Video Source",
+    "TikTok": "TikTok (TikTok support is coming soon)",
+    "Bilibili": "Bilibili (Bilibili support is coming soon)",
+    "Xiaohongshu": "Xiaohongshu (Xiaohongshu support is coming soon)",
+    "Local file": "Local file",
+    "Play Voice": "Play Voice",
+    "Voice Example": "This is an example text for testing speech synthesis",
+    "Synthesizing Voice": "Synthesizing voice, please wait...",
+    "TTS Provider": "Select the voice synthesis provider"
+  }
+}