I created a YouTube bot that creates educational videos based on Wikipedia articles as a fun little project.

To cut to the chase: https://www.youtube.com/channel/UCsLCKPmJXsDh90Eum8eMZXw

This is a project I have been working on over the past days. This is the first step in changing the future of readily accessible education in a natural, coherent, formatted manner. This project is just a sliver of the changes my team and I plan to make to the efficiency of education.

Essentially, what I've done is fully automate the process of creating educational videos based on the source. I've started out solely with wikipedia for this. All i have to do is paste any wikipedia article I want on a google docs, and my python script will take this article, narrate it using Google Cloud's Wavenet Text to Speech API for a more natural sounding voice, choose images using Amazon Comprehend Keyword Extraction API, as well as create a subtitle file, timestamped descriptions, and thumbnails. Once the script has created these, it will automatically upload both a full video on the topic and a summary video of the same topic on Youtube.

Of course, there are a few off-topic and some downright irrelevant images, but this is only the first step in the process. I also understand wikipedia may not be the most reliable source. However, I plan to expand sourcing past wikipedia eventually and create a better image selection algorithm.

I'd very much appreciate any subs or shares. Thank you!

Submitted July 17, 2020 at 04:36PM by FlamingGem https://ift.tt/2OCB1Cx