Verified project record

ATH-MaaS/Pixelle-Video

The project automates end-to-end short video creation from a single text prompt by combining scriptwriting, AI media generation, voiceover synthesis, and video composition. It is self-hosted via a web GUI and targets creators and general users who want automated video generation without requiring manual fine-grained editing control.

Project overview

Project type: Workflow · Image & Vision · Video
Use cases: Documents & Office · Video Creation · Audio & Speech
Deployment: Refer to project documentation
License: Apache-2.0

Best for

Creators and general users who want to generate short videos from a single text prompt without requiring prior editing experience.
Users who want to flexibly combine or replace image, video, TTS, and VLM components using ComfyUI workflows.

Key capabilities

Automatically writes video scripts, generates AI illustrations and videos, synthesizes voiceovers, adds background music, and composes the final video from a single text topic.
Generates intelligent explanatory scripts based on the input topic.
Supports various TTS solutions including Edge-TTS and Index-TTS, with support for voice cloning from reference audio.
Generates matching AI images and videos for sentences using models like WAN 2.1 or ComfyUI workflows.
Generates talking digital human videos.
Uploads reference videos and images to perform motion transfer.
Supports ComfyUI and RunningHub workflows to flexibly replace image, video, TTS, and VLM capabilities.
Allows users to upload their own photos and videos, using AI to analyze the materials and generate corresponding scripts.

Limitations and risks

Requires manual installation of ffmpeg for users not running the non-Windows integration bundle.
Cloud-based LLM and image generation services may incur higher fees depending on usage.

Getting started

Download and extract the latest Windows integration package. Double-click start.bat to launch the web interface, which automatically opens at http://localhost:8501. Configure the LLM API and image generation service in System Configuration to begin.
No coding is required to use the project. No Python, uv, or ffmpeg installation is needed for the Windows integration package.

Evidence and sources

README: 只需输入一个 **主题**，Pixelle-Video 就能自动完成： - ✍️ 撰写视频文案 - 🎨 生成 AI 配图/视频 - 🗣️ 合成语音解说 - 🎵 添加背景音乐 - 🎬 一键合成视频 **零门槛，零剪辑经验**，让视频创作成为一句话的事！
README: 🎬 Pixelle-Video —— AI 全自动短视频引擎
README: - ✅ **全自动生成** - 输入主题，自动生成完整视频 - ✅ **AI 智能文案** - 根据主题智能创作解说词，无需自己写脚本 - ✅ **AI 生成配图** - 每句话都配上精美的 AI 插图 - ✅ **AI 生成视频** - 支持使用 AI 视频生成模型（如 WAN 2.1）创建动态视频内容 - ✅ **直连模型 API** - 可直接调用…
README: - ✅ **2026-01-26**: 新增「动作迁移」模块，上传参考视频和图片进行动作迁移 - ✅ **2026-01-14**: 新增「数字人口播」和「图生视频」流水线，新增多语言 TTS 音色支持 - ✅ **2026-01-06**: 新增 RunningHub 48G 显存机器调用支持
README: - ✅ **2025-12-04**: 新增「自定义素材」功能，支持用户上传自己的照片和视频，AI 智能分析生成脚本

Reviewed at: 2026-07-17T10:29:50Z