An AI-powered, locally run video generator that transforms user prompts into short, engaging social media videos. This tool combines AI-generated scripts (via LLaMA 3.1), visual content (using Stable Diffusion and Pexels), and realistic text-to-speech (Edge TTS) to produce fully automated videos with subtitles. Videos are built from structured JSON scripts and assembled with synchronized images, audio, and effects. Use via command line or an optional Flask web server.
Due to the high computing demands, I unfortunately can't host the server publicly. However, if you're interested in how it works and what the results are, please look at the README.md on GitHub as it has a sample video I generated. Also, feel free to download the project and try it on your own computer if you have the specs for it (Just a warning that it requires a lot of computing power). The project is also open to any optimizations that could be made. This is my first project using local AI image generation so I'm sure there's room for improvement.
No followers yet
Once you ship this you can't edit the description of the project, but you'll be able to add more devlogs and re-ship it as you add new features!
Added a web server interface that allows users to easily connect with the backend to generate videos and then download them.