🎬 Pixelle-Video —— AI Fully Automated Short Video Engine

English | 中文

https://github.com/user-attachments/assets/a42e7457-fcc8-40da-83fc-784c45a8b95d

Just input a topic, and Pixelle-Video will automatically: - ✍️ Write video script - 🎨 Generate AI images/videos
- 🗣️ Synthesize voice narration - 🎵 Add background music - 🎬 Create video with one click

Zero threshold, zero editing experience - Make video creation as simple as typing a sentence!

🖥️ Web Interface Preview

📋 Recent Updates

• ✅ 2026-01-26: Added the Motion Transfer pipeline — upload a reference video and an image to transfer motion.
• ✅ 2026-01-14: Added "Digital Human" and "Image-to-Video" pipelines, multi-language TTS voices support
• ✅ 2026-01-06: Added RunningHub 48G VRAM machine support
• ✅ 2025-12-28: Configurable RunningHub concurrency limit, improved LLM structured data response handling
• ✅ 2025-12-17: Added ComfyUI API Key configuration, Nano Banana model support, API template custom parameters
• ✅ 2025-12-10: Built-in FAQ in sidebar, fixed edge-tts version to resolve TTS service instability
• ✅ 2025-12-08: Support multiple script split modes (paragraph/line/sentence), improved template selection with direct preview
• ✅ 2025-12-06: Fixed video generation API URL path handling with cross-platform compatibility
• ✅ 2025-12-05: Added Windows all-in-one package download, optimized image and video analysis workflows
• ✅ 2025-12-04: New "Custom Media" feature - upload your photos/videos with AI-powered analysis and script generation
• ✅ 2025-11-18: Parallel processing for RunningHub, added history page, batch video task creation support

✨ Key Features

• ✅ Fully Automatic Generation - Input a topic, automatically generate complete video
• ✅ AI Smart Copywriting - Intelligently create narration based on topic, no need to write scripts yourself
• ✅ AI Generated Images - Each sentence comes with beautiful AI illustrations
• ✅ AI Generated Videos - Support AI video generation models (like WAN 2.1) to create dynamic video content
• ✅ AI Generated Voice - Support Edge-TTS, Index-TTS and many other mainstream TTS solutions
• ✅ Background Music - Support adding BGM to make videos more atmospheric
• ✅ Visual Styles - Multiple templates to choose from, create unique video styles
• ✅ Flexible Dimensions - Support portrait, landscape and other video dimensions
• ✅ Multiple AI Models - Support GPT, Qwen, DeepSeek, Ollama and more
• ✅ Flexible Atomic Capability Combination - Based on ComfyUI architecture, can use preset workflows or customize any capability (such as replacing image generation model with FLUX, replacing TTS with ChatTTS, etc.)

📊 Video Generation Pipeline

Pixelle-Video adopts a modular design, the entire video generation process is clear and concise:

From input text to final video output, the entire process is clear and simple: Script Generation → Image Planning → Frame-by-Frame Processing → Video Composition

Each step supports flexible customization, allowing you to choose different AI models, audio engines, visual styles, etc., to meet personalized creation needs.

🎬 Video Examples

Here are actual cases generated using Pixelle-Video, showcasing video effects with different themes and styles:

📱 Extension Module Video Showcase

👤 AI Digital Avatar Korean-speaking AI Avatar

🖼️ Image-to-Video Animated Cartoon Video

💃 Motion Transfer Dancing Kitten

📱 Portrait Video Showcase

🌄 Documentary & Lifestyle – Default Template The Scenery Along the Journey	🔍 Cultural Deconstruction – Default Template Santa ID	🔭 Scientific Inquiry – Default Template Why Haven’t We Found Alien Civilizations Yet?
🌱 Personal Growth – Cloned Voice How to Level Up Yourself	🧠 Deep Thinking – Default Template Understanding Antifragility	🏯 History & Culture – Static Frame Zizhi Tongjian (Comprehensive Mirror for Aid in Governance)
☀️ Emotional Storytelling – Cloned Voice Winter Sunlight	📜 Novel Adaptation – Custom Script Doupo Cangqiong (Battle Through the Heavens)	🧬 Knowledge Explainer – Qwen Image Generation Essential Wellness Tips

🖥️ Landscape Video Showcase

💰 Side Hustle Money Making - Movie Template Side Hustle Money Making

🏛️ Historical Commentary - Custom Template Insights from Zizhi Tongjian

💡 Tip: All these videos are fully automatically generated by AI just by inputting a topic keyword, without any video editing experience required!

🚀 Quick Start

🪟 Windows All-in-One Package (Recommended for Windows Users)

No need to install Python, uv, or ffmpeg - ready to use out of the box!

👉 Download Windows All-in-One Package

1. Download the latest Windows All-in-One Package and extract it
2. Double-click start.bat to launch the Web interface
3. Browser will automatically open http://localhost:8501
4. Configure LLM API and image generation service in "⚙️ System Configuration"
5. Start generating videos!

💡 Tip: The package includes all dependencies, no need to manually install any environment. On first use, you only need to configure API keys.

Install from Source (For macOS / Linux Users or Users Who Need Customization)

Prerequisites

Before starting, you need to install Python package manager uv and video processing tool ffmpeg:

Install uv

Please visit the uv official documentation to see the installation method for your system:
👉 uv Installation Guide

After installation, run uv --version in the terminal to verify successful installation.

Install ffmpeg

macOS

brew install ffmpeg

Ubuntu / Debian

sudo apt update
sudo apt install ffmpeg

Windows - Download URL: https://ffmpeg.org/download.html - After downloading, extract and add the bin directory to the system environment variable PATH

After installation, run ffmpeg -version in the terminal to verify successful installation.

Step 1: Clone Project

git clone https://github.com/AIDC-AI/Pixelle-Video.git
cd Pixelle-Video

Step 2: Launch Web Interface

# Run with uv (recommended, will automatically install dependencies)
uv run streamlit run web/app.py

Browser will automatically open http://localhost:8501

Step 3: Configure in Web Interface

On first use, expand the "⚙️ System Configuration" panel and fill in: - LLM Configuration: Select AI model (such as Qwen, GPT, etc.) and enter API Key - Image Configuration: If you need to generate images, configure ComfyUI address or RunningHub API Key

After configuration, click "Save Configuration", and you can start generating videos!

💻 Usage

After opening the Web interface, you will see a three-column layout. Here's a detailed explanation of each part:

⚙️ System Configuration (Required on First Use)

Configuration is required on first use. Click to expand the "⚙️ System Configuration" panel:

1. LLM Configuration (Large Language Model)

Used for generating video scripts.

Quick Select Preset
- Select preset model from dropdown menu (Qwen, GPT-4o, DeepSeek, etc.) - After selection, base_url and model will be automatically filled - Click "🔑 Get API Key" link to register and obtain key

Manual Configuration
- API Key: Enter your key - Base URL: API address - Model: Model name

2. Image Configuration

Used for generating video images.

Local Deployment (Recommended)
- ComfyUI URL: Local ComfyUI service address (default http://127.0.0.1:8188) - Click "Test Connection" to confirm service is available

Cloud Deployment
- RunningHub API Key: Cloud image generation service key

After configuration, click "Save Configuration".

📝 Content Input (Left Column)

Generation Mode

• AI Generated Content: Input topic, AI automatically creates script
• Suitable for: Want to quickly generate video, let AI write script
• Example: "Why develop a reading habit"
• Fixed Script Content: Directly input complete script, skip AI creation
• Suitable for: Already have ready-made script, directly generate video

Background Music (BGM)

• No BGM: Pure voice narration
• Built-in Music: Select preset background music (such as default.mp3)
• Custom Music: Put your music files (MP3/WAV, etc.) in the bgm/ folder
• Click "Preview BGM" to preview music

🎤 Voice Settings (Middle Column)

TTS Workflow

• Select TTS workflow from dropdown menu (supports Edge-TTS, Index-TTS, etc.)
• System will automatically scan TTS workflows in the workflows/ folder
• If you know ComfyUI, you can customize TTS workflows

Reference Audio (Optional)

• Upload reference audio file for voice cloning (supports MP3/WAV/FLAC and other formats)
• Suitable for TTS workflows that support voice cloning (such as Index-TTS)
• Can listen directly after upload

Preview Function

• Enter test text, click "Preview Voice" to listen to the effect
• Supports using reference audio for preview

🎨 Visual Settings (Middle Column)

Image Generation

Determine what style of images AI generates.

ComfyUI Workflow
- Select image generation workflow from dropdown menu - Supports local deployment (selfhost) and cloud (RunningHub) workflows - Default uses image_flux.json - If you know ComfyUI, you can put your own workflows in the workflows/ folder

Image Dimensions
- Set width and height of generated images (unit: pixels) - Default 1024x1024, can be adjusted as needed - Note: Different models have different dimension limitations

Prompt Prefix
- Controls overall image style (language needs to be English) - Example: Minimalist black-and-white matchstick figure style illustration, clean lines, simple sketch style - Click "Preview Style" to test effect

Video