🎬 Nova Studio AI

Nova Studio AI is a production-grade, local, and offline non-linear video creation workspace. It orchestrates a modular pipeline of AI Agents (Script, Voice, Storyboard, Timeline Compiler) to convert raw ideas into complete editable video exports.

👉 Download the Standalone Windows Installer (.exe) from GitHub Releases

---

📖 Table of Contents

1. Problem Statement
2. Key Features
3. System Architecture
4. Folder Directory Layout
5. Getting Started & Installation
6. REST API Documentation
7. Plugin SDK Guidelines
8. Frequently Asked Questions (FAQ)
9. Roadmap
10. Contributing & License

---

💡 Problem Statement

Creating engaging short-form video content using AI models typically involves bouncing between multiple disjointed cloud services: LLM script generators, TTS speech synthesizers, image generation interfaces, and traditional NLE timeline editors. This workflow suffers from:

• High Latency & Costs: Relying on expensive cloud APIs.
• Privacy Risks: Storing drafts and inputs on remote servers.
• Disjointed Automation: No smooth path from text script to synced subtitle audio tracks and clips timelines.

Nova Studio AI solves this by packaging the entire pipeline locally, routing calculations offline to local models (Ollama, Kokoro TTS, ComfyUI) inside an integrated non-linear timeline workspace.

---

🌟 Key Features

• Decoupled Event Bus: Modules publish state changes to a thread-safe broker, enabling custom plugin extensions without touching core logic.
• Non-Linear Editor (NLE): Interactive lanes supporting splits, duplicate clones, ripple deletes, and version history rollback snapshots.
• Vocal Synthesis & Lip Sync: Kokoro TTS engine generating viseme mouth shape mappings and subtitle karaoke grids.
• Database & Cache Auto-optimizer: Reclaim unused SQLite space and scan asset file sizes automatically.
• REST API Gateway: Background daemon HTTP server listening on port 9000 with webhook dispatcher queues.

---

⚙️ System Architecture

The studio leverages a fully event-driven architecture to communicate across layers:

graph TD
    UI[Streamlit UI Dashboard] -->|1. Triggers Action| GW[APIGateway]
    GW -->|2. Broadcasts Event| EB[Event Bus]
    
    EB -->|3. Callback| AUTO[Automation Engine]
    EB -->|4. Post JSON| WH[Webhook Dispatcher]
    EB -->|5. Push logs| REST[REST API server:9000]
    
    UI -->|6. Maintain state| DB[(SQLite Database)]
    UI -->|7. Security layers| SEC[Security key manager]

Loading

---

📂 Folder Directory Layout

nova-studio/
├── core/
│   ├── api/            # EventBus, APIGateway, RESTServer, TaskQueue, Webhooks
│   ├── comfy/          # ComfyUI Connectors & Workflow Managers
│   ├── timeline/       # NLE Timeline cuts, snapshots versions
│   ├── generation/     # Structured prompts, characters consistency profile
│   ├── audio/          # Subtitle karaoke compilers, sidechain ducking, TTS
│   ├── database/       # SQLite storage indexing & tables
│   ├── logger/         # Rotating action logger
│   └── plugins/        # swappable provider script loaders
├── installer/          # One-Click Bootstrap Installer suite for Windows
├── plugins/            # folder-based SDK plugins (e.g. sample_plugin)
├── tests/              # unit test suites
├── app.py              # Streamlit dashboard
├── requirements.txt    # dependencies list
└── README.md           # user guide

---

🚀 Getting Started & Installation

Option A: Professional Windows Installer compilation (Developers)

To compile a single executable wizard (NovaStudioAI_Setup.exe) similar to VS Code:

1. Navigate to the installer_project/ folder.
2. Run the build command:

   .\build.bat

   Requires PyInstaller and Inno Setup 6 compiler.

Option B: One-Click Bootstrap Script (Windows Recommended)

If you are on Windows 10 or 11, you can set up the entire project automatically:

1. Navigate to the installer/ folder.
2. Right-click setup.bat and select "Run as administrator".
3. Once the setup completes, double-click the new "Launch Nova Studio AI" shortcut on your Desktop!

Option C: Manual Installation

If you prefer a manual setup or are on Linux:

Prerequisites

• Python 3.10 or 3.11 installed.
• FFmpeg installed and added to your system environment variables.

1. Installation

Clone the repository and install requirements inside a virtual environment:

git clone https://github.com/<your-username>/nova-studio-ai.git
cd nova-studio-ai

# Set up virtual environment
python -m venv venv
source venv/bin/activate  # Or venv\Scripts\activate on Windows

# Install dependencies
pip install -r requirements.txt

2. Run the Dashboard

Start the application server using:

streamlit run app.py

Open http://localhost:8501 in your browser.

---

🌐 REST API Documentation

The background daemon HTTP server runs on http://127.0.0.1:9000/api.

Endpoints Details

Method	Endpoint	Description	Example Response
GET	/api/system	Queries OS, GPU, and disk statistics.	{"gpu_name": "NVIDIA RTX 4080", ...}
GET	/api/projects	Lists all project metadata logs.	[{"id": "proj_123", "name": "Untitled"}]
GET	/api/history	Fetches Event Bus dispatched history logs.	[{"id": "evt_abc", "type": "Project Saved"}]

---

🔌 Plugin SDK Guidelines

Developers can register folder-based plugins under plugins/<name>/.

Required Files

• plugin.json: Metadata, categories, and requested sandbox permissions.
• main.py: SDK subclass implementation.

Example plugin.json

{
    "id": "sample_plugin",
    "name": "Sample Transitions",
    "version": "1.0.0",
    "category": "exporter",
    "permissions": {
        "filesystem": true,
        "network": false
    }
}

---

❓ Frequently Asked Questions (FAQ)

1. Why is ComfyUI status marked offline?

ComfyUI server must be running locally on port 8188. If offline, the studio samplers run in mock simulation fallback.

2. How can I optimize the database size?

Go to the System Diagnostics tab in the UI sidebar and click Re-Optimize SQLite database to run VACUUM.

---

🗺️ Roadmap

• AI Voice Cloning: Integrate local model checkpoints cloning audio inputs.
• AI Lip Sync: Generate mouth-shape transformations matched to dialogue wavs.
• Distributed Rendering: Coordinate multiple rendering nodes across networks.

---

🤝 Contributing & License

For guidelines, please check CONTRIBUTING.md.

This project is licensed under the MIT License.