An approach to reverse engineer high-frequency media persuasion

🆓 Free* video and short content creation with AI 🆓

Creator	GPT Model	Estimated Monthly Earnings (€)	Source (estimated)
RayVentura	ShortGPT	€0.45 - €7	Socialblade (@AskRedditors)
Su77ungr	FreeShortGPT	€24 - €389	Socialblade (@UnsoundFlash)

Installation and Use

*

The Main branch relies on the paid OpenAI API to work. This has been resolved (not released yet❗) by implementing LlamaCpp for locally hosted Langchain agents instead. For setup refer to the main repo and installation guide

Upload Automation

1. Authentication (client-secrets.json from Google API console)

• Head to the Google API Console at https://console.cloud.google.com/ and create a project.

• In the marketplace (search bar), find 'YouTube v3 API' and enable it.

• After redirect, click on Oauth Screen and create.

• Add google.com as the authorized domain.

-> Add scopes for the YouTube v3 API.

• Add the channel's email as the test user (in most cases, the email the channel was registered with).

• Create Credentials (OAuth Client ID) in the Credentials Menu and select Web App.

• For Authorized Redirect URIs, use:

  1. http://localhost:8080
  2. http://localhost:8080/
  3. http://localhost:8090
  4. http://localhost:8090/

• Finish up with downloading the .JSON file and put it inside this repository's directory.

Note: For more information about the client_secrets.json file format, please visit: api-client-library/python/guide/aaa_client_s

2. Oauth2.json (within the repo itself)

• Once a valid client_secrets.json is present in the directory, run python3 upload.py and follow further instructions. After successful authentication, an oauth2.json file will be created inside the directory.

Note: You can avoid quota bottlenecks by stacking up projects in the google console and repeat steps before

3. Automation

Once valid oauth2 credentials exist in the directory you can let this run 100% on autopilot.

• ultra.py is the main script to direct the video production pipeling
• upload.py sets the backbone for youtube's v3 api to upload to YouTube without further authentification
• final_upload.py combines the finsished product and uploads it to YouTube (title, description, tags get parsed automatically)

Just craft your own bash script or use automate.sh

Upgrades

• access YouTube as provider of video assets, surpassing stock footage by large
• avoids bans by using download streams with PyTube
• avoids YouTube v3 API's rate limit and auth
• avoids Pexels hard rate limit

Features

Inside api_utils functions provide utility for working with different APIs. Files: image_api.py, pexels_api.py, youtube_api.py and eleven_api.py. We added youtube_api.py to source video assets directly from YouTube. Feel free to modify the _generateVideoUrls function for the hierachy of video asset sources.

• search_videos_YouTube(query_string)

  The search_videos_YouTube function takes a query string as input and searches for a video on YouTube based on that query. It returns the URL of the first search result if found, or None if no video is found.

  Integration of shortGPT.api_utils.youtube_api, which serves as an additional source for gathering footage based on the queried keyword. This is particularly useful when dealing with niche / comedy / meme topics where stock footage is not available. If nothing matches we backtrack to the pexels API.

• triage_videos_Youtube(expected_score_parameter) ❗not released yet

Demo (average view duration: 143% @ 10k views)

demo_new shows the accuracy of the newly added youtube query policy without further guidance, backtesting or content analysis. This can be improved by adding a content triage based on Clip2 and transcript analysis.