YouTube Cleaner

Inspiration

YouTube often demonetizes videos that have swearing and derogatory words in them. As users of YouTube, we know the struggles that many YouTubers have with trying to filter themselves in videos to avoid getting demonetized. It is also a pain to have to re-edit your video to take out the cursing when you just finished editing it, losing money because you are now behind on the upload schedule.

What it does

Upload a video you just filmed or a YouTube URL, and it will edit out all curse words. Download the video and resubmit it on YouTube without having to edit it yourself!

How we built it

We used Python and the Flask library to build the webserver. Assembly AI and Google Cloud Storage APIs are used to process the YouTube videos and edit out the profanity. Assembly AI is utilized to transcript the video and have a word search filter to mute the appropriate parts of the video.

Challenges we ran into

There were a lot of different parts working together like the Flask web server and Assembly AI, so it was a challenge to combine everyone’s work together. We also needed to learn how to make a web server with Flask and how to transcribe with Assembly AI.

Accomplishments that we're proud of

None of us knew anything about AI so being able to use Assembly AI to transcribe videos and filter out the profanity is exciting. We also learned a lot about Flask, specifically extensions, routes, and services.

What we learned

We learned how to build production-ready apps with Flask using extensions, routes, and services. Additionally, we learned a lot about using Assembly AI like how to transcribe videos and apply a word filter on videos. It was interesting to learn more about MP4 and MP3 processing.

What's next for YouTube Cleaner

Next for YouTube Cleaner is to consider making it a Google Chrome extension implementation to give users more convenience. We also need to update swear words and derogatory words as the English language always comes up with more words and slang over time.