{"id":49444924,"url":"https://github.com/toki-plus/ai-video-workflow","last_synced_at":"2026-04-29T21:40:10.391Z","repository":{"id":328024630,"uuid":"1113998904","full_name":"toki-plus/ai-video-workflow","owner":"toki-plus","description":"全自动AI原生视频生成工作流，集成文生图(LibLib)/图生视频(即梦)/文生音乐(即梦)和AI提示词生成(豆包)，一键创作AIGC短视频。generative-ai, text-to-video, image-to-video, text-to-music, aigc, automation, pyqt5.","archived":false,"fork":false,"pushed_at":"2026-01-12T09:57:02.000Z","size":10782,"stargazers_count":12,"open_issues_count":1,"forks_count":5,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-01-12T18:48:23.334Z","etag":null,"topics":["ai-content-creation","aigc","automation","bot","content-creation","ffmpeg","llm","pyqt5","python","short-video","tiktok","video-processing"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/toki-plus.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-12-10T18:52:50.000Z","updated_at":"2026-01-12T09:57:05.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/toki-plus/ai-video-workflow","commit_stats":null,"previous_names":["toki-plus/ai-video-workflow"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/toki-plus/ai-video-workflow","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/toki-plus%2Fai-video-workflow","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/toki-plus%2Fai-video-workflow/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/toki-plus%2Fai-video-workflow/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/toki-plus%2Fai-video-workflow/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/toki-plus","download_url":"https://codeload.github.com/toki-plus/ai-video-workflow/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/toki-plus%2Fai-video-workflow/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32445547,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-29T20:22:27.477Z","status":"ssl_error","status_checked_at":"2026-04-29T20:22:26.507Z","response_time":110,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.6:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-content-creation","aigc","automation","bot","content-creation","ffmpeg","llm","pyqt5","python","short-video","tiktok","video-processing"],"created_at":"2026-04-29T21:40:08.296Z","updated_at":"2026-04-29T21:40:10.378Z","avatar_url":"https://github.com/toki-plus.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# AI Video Workflow: 全自动 AI 原生视频生成工作流\n\n[简体中文](./README.md) | [English](./README_en.md)\n\n[![GitHub stars](https://img.shields.io/github/stars/toki-plus/ai-video-workflow?style=social)](https://github.com/toki-plus/ai-video-workflow/stargazers)\n[![GitHub forks](https://img.shields.io/github/forks/toki-plus/ai-video-workflow?style=social)](https://github.com/toki-plus/ai-video-workflow/network/members)\n[![MIT License](https://img.shields.io/badge/License-MIT-green.svg)](https://choosealicense.com/licenses/mit/)\n[![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](https://github.com/toki-plus/ai-video-workflow/pulls)\n\n**`AI Video Workflow` 是一款免费、开源的桌面应用程序，它将多个顶尖的AI模型（文生图、图生视频、文生音乐）串联成一个全自动的AI原生视频创作流水线。**\n\n你是否想创作引人注目的AI短视频，却被繁琐的平台切换、复杂的参数调整和枯竭的创作灵感所困扰？本项目旨在将AI视频创作的整个过程简化为几次点击，让任何人都能轻松生成具有独特视觉风格和定制化配乐的短视频。\n\n\u003cp align=\"center\"\u003e\n  \u003ca href=\"https://www.bilibili.com\" target=\"_blank\"\u003e\n    \u003cimg src=\"\" alt=\"点击观看B站演示视频（暂未录制）\" width=\"800\"/\u003e\n  \u003c/a\u003e\n  \u003cbr\u003e\n  \u003cem\u003e(点击图片跳转到 B 站观看高清演示视频)\u003c/em\u003e\n\u003c/p\u003e\n\n---\n\n## ✨ 核心功能\n\n这不仅是一个工具，更是一个完整的 AIGC 创作生态系统：\n\n-   **🤖 全自动AI创作流水线**:\n    -   **文生图 (Text-to-Image)**: 对接 **LibLibAI** 平台，支持丰富的 Checkpoint、LoRA 模型和参数，将您的想法变为精美图像。\n    -   **图生视频 (Image-to-Video)**: 调用火山引擎 **即梦（Jimeng）I2V** 模型，为静态图片赋予生命，生成平滑自然的动态视频。\n    -   **文生音乐 (Text-to-Music)**: 接入火山引擎 **即梦（Jimeng）音乐** 模型，通过文本描述（如风格、情绪、乐器）即可生成独一无二的背景音乐。\n    -   **自动合成 (Automatic Merging)**: 使用强大的 **FFmpeg** 引擎，将生成的视频画面与背景音乐无缝合成为最终的成品视频。\n\n-   **💡 AI驱动的灵感引擎**:\n    -   内置基于 **豆包（Doubao）大模型** 的提示词生成器。\n    -   提供“美女”、“Labubu”等多种预设主题，只需勾选想要的风格（如“沙滩”、“健身房”或“糖果系”、“魔法系”），即可一键生成全套专业的图片、音乐提示词和爆款标题、标签。\n\n-   **🎨 直观的图形化界面 (GUI)**:\n    -   **三步式工作流**: 清晰地将创作过程分为“文生图 → 图生视频 → 文生音乐与合成”三个阶段，每一步的进展和结果都一目了然。\n    -   **集中式参数管理**: 在统一的界面中配置所有AI模型的参数，无需在多个网页或应用间切换。\n    -   **实时媒体预览**: 生成的图片和视频会直接在界面中展示和播放，方便您即时评估效果。\n    -   **历史记录与导航**: 支持在多张生成的图片之间轻松切换，方便您选择最满意的一张进入下一步。\n\n## 📸 软件截图\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"./assets/cover_software01.png\" alt=\"软件主界面\" width=\"800\"/\u003e\n  \u003cbr\u003e\n  \u003cem\u003e软件主界面：提示词生成部分。\u003c/em\u003e\n\u003c/p\u003e\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"./assets/cover_software02.png\" alt=\"软件主界面\" width=\"800\"/\u003e\n  \u003cbr\u003e\n  \u003cem\u003e软件主界面：图像参数部分。\u003c/em\u003e\n\u003c/p\u003e\n\n\u003ctable align=\"center\"\u003e\n  \u003ctr\u003e\n    \u003ctd align=\"center\" valign=\"top\"\u003e\n      \u003cimg src=\"./assets/cover_demo_picture.png\" alt=\"生成示例：生成图像\" width=\"390\"/\u003e\n      \u003cbr /\u003e\n      \u003cem\u003e生成示例：生成图像。\u003c/em\u003e\n    \u003c/td\u003e\n    \u003ctd align=\"center\" valign=\"top\"\u003e\n      \u003cimg src=\"./assets/cover_demo_video.gif\" alt=\"生成示例：最终视频\" width=\"390\"/\u003e\n      \u003cbr /\u003e\n      \u003cem\u003e生成示例：最终视频。\u003c/em\u003e\n    \u003c/td\u003e\n  \u003c/tr\u003e\n\u003c/table\u003e\n\n## 🚀 快速开始\n\n### 系统要求\n\n1.  **Python**: 3.8 或更高版本。\n2.  **FFmpeg**: **必须**安装 FFmpeg 并将其添加到系统环境变量中。\n    -   请访问 [FFmpeg 官网](https://ffmpeg.org/download.html) 查看安装教程。\n    -   检查是否安装成功：打开终端或命令提示符，输入 `ffmpeg -version`。\n3.  **API Keys**:\n    -   **豆包（Doubao） API Key**: 用于提示词生成。\n    -   **LibLibAI Access Key \u0026 Secret Key**: 用于文生图。\n    -   **火山引擎（即梦）Access Key \u0026 Secret Key**: 用于图生视频和文生音乐。\n\n### 安装与启动\n\n1.  **克隆本仓库：**\n    ```bash\n    git clone https://github.com/toki-plus/ai-video-workflow.git\n    cd ai-video-workflow\n    ```\n\n2.  **创建并激活虚拟环境 (推荐)：**\n    ```bash\n    python -m venv venv\n    # Windows 系统\n    venv\\Scripts\\activate\n    # macOS/Linux 系统\n    source venv/bin/activate\n    ```\n\n3.  **安装依赖库：**\n    ```bash\n    pip install -r requirements.txt\n    ```\n\n4.  **配置 API Keys:**\n    -   **强烈建议**通过设置系统环境变量来配置密钥，程序会自动读取：\n        - `DOUBAO_API_KEY`\n        - `LIBLIB_AK`, `LIBLIB_SK`\n        - `JIMENG_AK`, `JIMENG_SK`\n    -   或者，您也可以在软件启动后，在 \"API 密钥\" 标签页中手动输入。\n\n5.  **运行程序：**\n    ```bash\n    python ai_video_workflow.py\n    ```\n\n## 📖 使用指南\n\n1.  **第一步：配置与准备**\n    -   启动软件，在左侧的 \"API 密钥\" 标签页中确认所有密钥已填写正确，点击“保存当前参数”应用。\n    -   切换到 \"提示词生成\" 标签页，选择一个您感兴趣的主题（如“美女”），勾选几个场景，然后点击“生成提示词”。\n    -   在下方生成的表格中，选择最喜欢的一行，点击“应用选中行提示词”。\n\n2.  **第二步：文生图**\n    -   参数会自动填充到“图像参数”和“音视频参数”标签页，您也可以手动修改。\n    -   在右侧工作流面板，点击“开始生成图片”。等待片刻，生成的图片将显示在预览区。\n    -   您可以多次生成，并通过“上一张”/“下一张”按钮选择最满意的图片。\n\n3.  **第三步：图生视频**\n    -   确认已选中满意的图片后，点击“生成视频”。程序会将该图片发送到AI模型进行处理。\n    -   处理完成后，生成的无声视频会自动在预览区循环播放。\n\n4.  **第四步：文生音乐与合成**\n    -   点击“合成最终视频”。程序将使用“音视频参数”中的音乐提示词生成配乐，并与视频合并。\n    -   任务完成后，最终的带配乐视频将在预览区播放，并保存在 `output` 文件夹中。\n\n5.  **完成！**\n    -   点击“全部重来”可以清空当前状态，开始一次全新的创作。\n\n---\n\n\u003cp align=\"center\"\u003e\n  \u003cstrong\u003e技术交流，请添加：\u003c/strong\u003e\n\u003c/p\u003e\n\u003ctable align=\"center\"\u003e\n  \u003ctr\u003e\n    \u003ctd align=\"center\"\u003e\n      \u003cimg src=\"./assets/wechat.png\" alt=\"微信二维码\" width=\"200\"/\u003e\n      \u003cbr /\u003e\n      \u003csub\u003e\u003cb\u003e个人微信\u003c/b\u003e\u003c/sub\u003e\n      \u003cbr /\u003e\n      \u003csub\u003e微信号: toki-plus (请备注“GitHub 定制”)\u003c/sub\u003e\n    \u003c/td\u003e\n    \u003ctd align=\"center\"\u003e\n      \u003cimg src=\"./assets/gzh.png\" alt=\"公众号二维码\" width=\"200\"/\u003e\n      \u003cbr /\u003e\n      \u003csub\u003e\u003cb\u003e公众号\u003c/b\u003e\u003c/sub\u003e\n      \u003cbr /\u003e\n      \u003csub\u003e获取最新技术分享与项目更新\u003c/sub\u003e\n    \u003c/td\u003e\n  \u003c/tr\u003e\n\u003c/table\u003e\n\n## 📂 我的其他开源项目\n\n-   **[Netease Downloader](https://github.com/toki-plus/netease-downloader)**: 一款优雅、功能丰富的网易云音乐下载器，支持无损/高品质音质、歌单/专辑批量下载、扫码登录和自动写入ID3元数据。\n-   **[AI-Trader-For-MT4](https://github.com/toki-plus/ai-trader-for-mt4)**: 革命性开源框架，将大语言模型（LLM）转变为能在MetaTrader 4（MT4）平台上进行自主交易的AI代理。\n-   **[Auto USPS Tracker](https://github.com/toki-plus/auto-usps-tracker)**: 专为跨境电商卖家设计的高效USPS批量物流追踪器，支持防屏蔽抓取并生成精美Excel报告。\n-   **[AI Mixed Cut](https://github.com/toki-plus/ai-mixed-cut)**: 一款颠覆性的AI内容生产工具，通过“解构-重构”模式将爆款视频解构成创作素材库，并全自动生成全新原创视频。\n-   **[AI Highlight Clip](https://github.com/toki-plus/ai-highlight-clip)**: 一款AI驱动的智能剪辑工具，能够全自动地从长视频中分析、发现并剪辑出多个“高光时刻”短视频，并自动生成爆款标题。\n-   **[AI TTV Workflow](https://github.com/toki-plus/ai-ttv-workflow)**: 一款AI驱动的文本转视频工具，能将任意文案自动转化为带有配音、字幕和封面的短视频，支持AI文案提取、二创和翻译。\n-   **[AB Video Deduplicator](https://github.com/toki-plus/AB-Video-Deduplicator)**: 通过创新的“高帧率抽帧混合”技术，从根本上重构视频数据指纹，以规避主流短视频平台的原创度检测和查重机制。\n-   **[Video Mover](https://github.com/toki-plus/video-mover)**: 一个强大的、全自动化的内容创作流水线工具。它可以自动监听、下载指定的博主发布的视频，进行深度、多维度的视频去重处理，并利用AI大模型生成爆款标题，最终自动发布到不同平台。\n\n## 🤝 参与贡献\n\n欢迎任何形式的贡献！如果你有新的功能点子、发现了Bug，或者有任何改进建议，请：\n-   提交一个 [Issue](https://github.com/toki-plus/ai-video-workflow/issues) 进行讨论。\n-   Fork 本仓库并提交 [Pull Request](https://github.com/toki-plus/ai-video-workflow/pulls)。\n\n如果这个项目对你有帮助，请不吝点亮一颗 ⭐！\n\n## 📜 开源协议\n\n本项目基于 MIT 协议开源。详情请见 [LICENSE](LICENSE) 文件。\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftoki-plus%2Fai-video-workflow","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftoki-plus%2Fai-video-workflow","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftoki-plus%2Fai-video-workflow/lists"}