About
DeepfakeOS™ is a Linux toolkit for creating deepfakes and other synthetic media. DeepfakeOS™ contains a selection of open source tools for creating synthetic video, audio, and text. The goal of DeepfakeOS™ is to lower the barrier of entry to this cutting edge technology by providing a one stop shop preloaded with the latest developments in open source deepfake technology.
System Requirements
Due to the GPU intensive nature of these tools it is recommended that you have a Nvidia 1080ti graphics card or better. It is highly recommended to install DeepfakeOS™ on a physical host as running in a virtual machine is not officially supported.
Download
Please log in to request access to DeepfakeOS™.
Installation Guide
1. Download and install Ubuntu 24.04. During installation it is recommend that you choose to install proprietary drivers and media formats when prompted.
2. Verify that you have a working Nvidia driver by running the command "nvidia-smi".
3. Download the DeepfakeOS™ toolkit and extract it.
4. Use the command "./setup1.sh" to run the first setup script. Do not run as root or use "sudo" as this will prevent the script from updating path variables in your ".bashrc". The script will prompt you for your password and once complete will restart your computer.
5. Use the command "./setup2.sh" to run the second setup script. Do not run as root or use "sudo" as this will prevent the script from updating path variables in your ".bashrc". The script will prompt you for your password and once complete will restart your computer.
The installation is now complete and DeepfakeOS™ is ready for use.
Tool List
For detailed instructions please refer to the websites or GitHub pages of the individual tools.
Deep Face Lab is a program for training deepfake models and using them to make pre-recorded videos. Its GitHub page has been taken down, but you can still find a detailed guide for it here.
Deep Face Live uses models trained with Deep Face Lab to make real time deepfakes. It comes with several pre-made models and its GitHub page can be found here.
Live Portrait is a tool for copying facial expressions from a source video onto a target image or video. Its GitHub page can be found here.
Ollama is a framework for locally hosted LLMS which are capable of uncensored text generation. Read more about it here.
RVC is an open source tool for voice cloning with a web interface. Its GitHub page is available here.