Problem Description

I'm looking to use open-source libraries to edit and stitch together images, audio, text, and video into composite videos similar to how Creatomate or Bannerbear APIs work. I've been able to successfully combine audio clips with images and multiple video clips into one using some combination of MoviePy and PIL but I can't seem to get the text effects to work in Replit despite trying numerous approaches.

Acceptance Criteria

As a user, I can programmatically combine any combination of images, text, audio, and video into a final rendered video. Utilizing effects such as pan & zoom, text animations, and positioning, I can create social media posts with effects similar to what Creatomate or Bannerbear offer. A great sample workflow would be to take 3-4 images, overlay audio on them, create a simple zoom effect on each image and transition effect between images, add tiktok style text effects to the image, and stitch everything together into one video file. Major bonus if there's a programmatic way to get the text to match the audio being spoken at the exact timing of each word.

Technical Details

I think my issues have been around some dependencies I can't get past but I honestly don't know for sure. I've tried doing this with ImageMagick but that doesn't seem to work in Replit. I've alsy tried openCV and other approaches but get stuck with the text manipulation.

Link to Project

Note, the project below is a subset of a bigger project but I tried cleaning up to just focus on this problem.
https://replit.com/@newmediamavs/VideoEditingTest