Exciting news! I've just released a video on Wunjo AI, fulfilling the requests many of you had. Just a quick reminder: Wunjo AI on GitHub is a local application that operates seamlessly even on less powerful computers. It empowers you to create content of any duration, a capability that web services inherently lack. Unlike web-based platforms, Wunjo AI can efficiently process hour-long videos, and the best part is it can run directly from a flash drive. Recently, I added a filter to enhance safety by blocking 18+ content in deepfakes. The application is entirely free and open-source, utilizing your computer's resources rather than relying on external servers.
I frequently get inquiries about creating GPU builds on Windows, with some facing challenges using briefcase. Now, there's a video answering all those questions. Additionally, I've included information on how to apply Wunjo AI modules in your projects using POST/GET requests, demonstrated with Postman examples. In essence, it's valuable information for developers, and I believe you'll find it interesting. If so, please show your support.
I created Wunjo AI to simplify the process of media content creation for authors and to explore general artificial intelligence for content generation. Version v2 introduces the idea of combining all modules with a language model. Here's a brief history of the project's evolution:
- Initially launched with speech synthesis only.
- Later added one-click deepfake face and lip syncing, speech cloning, and object removal from videos.
- Evolved to allow object and video alterations through textual requests.
- Introduced voice cloning from songs, improved video changes with text, and one-click text removal from videos.
Top comments (0)