A Python program that allows users to create TikTok-style videos by combining two video sources: one for gameplay (bottom) and one for a talking person (top). Customize with features like tracking the speaker, clip selection, transcription, and text editing.
- Combine Two Videos: Merge a gameplay video at the bottom and another video at the top for a TikTok-style video.
- Talking Head Tracking: Optionally track the person in the top video or keep the view static.
- Clip Selection: Choose specific sections of each video to include in the final edit.
- Multiple Clips: Add multiple clips for a dynamic final video.
- Transcription and Text Editing: Automatically generate a transcript with options to edit and style text for subtitles or annotations.
- Python 3.7+
- FFmpeg (for video processing)
- OpenCV (for face tracking)
- Whisperx (For transcription)
https://x.com/i/status/1830399524310770062
All secrete keys and access tokens uploaded accidentally to this repository have been deleted, to use this respository you will have to download the requirements by yourself as this was my first project and was hard to keep track of every library used. It took me a total of 1 year and 2 months to "complete" this project, I have decided to move on to creating something similar but all web based with more features.