DISCLAIMER: The product has already been started but the developer had personal issues and was not able to finish. The project is described below, and what has been finished already will be mentioned later.

Project:
We need a python project that imports open-source AI models and has the user upload a text script, a vocal sample, and a video. We then use those files with the AI models to combine them all into a single video. Here’s how it works:

– Use Tortoise-TTS to convert the script and vocal sample into speech based on the vocal sample’s voice and what is said in the script.
– We then use that AI created speech, and the user-uploaded video, with another re-talking AI model that will make the user-uploaded video seem like the person in the video is making the speech. We know what open-source model to use here, but we will privately message you this.
– All the above outputs a single video file that the user can download from the GUI
– We need this Python program to be a Windows executable that can be installed easily using an installer (can be done with PyInstaller)

What has been finished:
– The GUI is finished where the user can upload the files, with a "loading page" as well as a video preview page.
– We already have the part where the vocal sample and text file is uploaded and processed by Tortoise-TTS.

What needs to be done, are a bit more work with the text-to-speech, using the generated speech with the video to move the lips in the video, and making it easily installable on any windows device that has NVidia graphics cards.

**We do not need someone with experience in AI**
We just need to import AI models that are already available and open source, so only Python experience is needed.

Budget: $2,500

Posted On: February 09, 2024 23:35 UTC
Category: Desktop Software Development
Skills:Python

Country: United States

click to apply