The second automated Youtube video maker I made. Only now, it's time for memes.
It allows users to easily create a video like this one, which is made by using only this tool: https://youtu.be/ifVO8gNJTb8
Offers a clickable GUI where you can select multiple image and GIF memes. It automatically reads all text in the memes and lets you read them out, while revealing them part-by-part.
This means a video can be created in a matter of seconds or minutes depending on length, no editing required.
first.mp4
second.mp4
final-video.mp4
-
Node v14 (use
fnm
ornvm
to easily switch between versions) -
Install
ffmpeg
, it's a pre-requisite
Set environment variables in {client,server}/.env
(see .env.example
for reference)
Needs AWS keys for S3 and DynamoDB. Needs Google Cloud credentials for the Image-to-Text OCR.
Uses two AWS S3 buckets, and one dynamoDB table. See src/server.ts
Build and run the server production build
$ npm i
$ npm run build
$ npm start
Build the client
$ npm run client-build
# Now client/build contains a static build of the site
Serve the builded directory on netlify or any web server