🎬 VideoGen is an AI-powered backend system designed to automate video creation from text input. It provides a set of robust APIs that allow users to generate, edit, and manage personalized videos directly from the web.
Watch Demo Video:
Frontend Repository: https://github.com/DINH1022/videogen_frontend
- Secure JWT-based login and registration.
- Role-based access to protect API endpoints.
- Generate video scripts using Gemini AI from short prompts.
- Editable script workspace for full customization.
- Convert scripts into high-quality voiceovers via GroqAI TTS.
- Supports multiple voice styles and accents.
- Generate illustrations using Gemini Vision AI.
- Combine with uploaded images to enrich storytelling.
- Drag-and-drop support for multiple images.
- Reorder and preview images before creating video.
- Merge audio + images into synchronized MP4 videos.
- Automatically compute durations and transitions.
- Visual drag-and-drop timeline editor for images/audio.
- Live preview with duration indicators.
- Save drafts to continue editing later.
- View, update, or replay past projects.
- Choose different styles/themes for the final video.
- Add background music, effects, and transitions.
- Export in MP4 format and resolutions up to 4K.
- Manual sharing to platforms like YouTube, TikTok, with future social API support.
- Track engagement data like views, likes, watch time via dashboard.
Backend Stack
- Language:
Java 17 - Framework:
Spring Boot 3.2.x - Tools: Spring Security (JWT), Spring Web, JPA, MapStruct, Lombok
- DB:
PostgreSQL - Build Tool:
Maven
Frontend Compatibility
- ReactJS + Vite + Material UI (see frontend repo)