Faceless Dancer is building a World Model. Not an extension or a fine tuning of an existing generative model, but a multi modal core model built to understand and infer the physics of rhythm using existing open source models and encoders as its appendages.
Development continues on Faceless projects. Updates are coming soon. In the meantime, explore our available features : DataCut - an advanced video editing app with dataset focus, Klusterd - A unique new media discovery platform, Dance Station - open source AI creator tools
Thousands of new Donghua, Anime, Movies, and Steam Games have been added to our catalogue! Remember, if you think there is a media item missing, you can add it through the menu. Try to include as many fields as possible for maximum utility!
10 new Effects and 10 new Transitions have been added to DataCut - our open source video editor. Updated build is available in the repo : https://t.co/5BxNTJ8o7F
Make an account, start defining the specific concepts that you enjoy or dislike, find and be found by users who recognize that specific concept and, together, find the works that have something that matters to YOU.
Introducing Klusterd (https://t.co/NRqH5sbecE - @klusterd_dev ). Klusterd is not a review or rating site - it is a Content Discovery Platform. In a world where content is created more rapidly than ever, Klusterd will be a tool that helps you find and define what matters to YOU
Right now Klusterd has a huge collection of movies, pc games, anime, and donghua, but the collection is going to keep expanding into every content category that is rapidly growing.
Adapter and model building continues. But also working on something big that seeks to solve some problems that AI's influence on total available content has exacerbated. Results of a nifty side project coming soon.
Follow along to get the latest on updates to DataCut, our upcoming melodic voice changing adapters, progress on Dance World, and new features and tools!
Introducing DataCut 0.7-beta, a lightweight and powerful video editing app for Windows with auto-caption, voice changing, and image/video/music/sound effect generation that run directly on your local machine; and frame-specific features to make dataset editing easier
The best part is that auto-caption doesn't require a GPU at all, just 2GB of system RAM, and the rest of the generative features only require a GPU with 10GB of VRAM.
https://t.co/5BxNTJ7Qi7