mestvnvo/Bop-Dropper

Music Recommender System: Dropping some EDM bops near you!

2

stars

63

commits

Python

primary language

Jul 9, 2025

updated

www.bopdropper.com

README

Bop Dropper

Dropping some EDM bops near you!

Paste in any song from this playlist! https://open.spotify.com/playlist/2c8qTVOuNvJIl8UepuzHBb?si=8e68fd3da0b44dfe

Notes

Why?

I recently started DJing and created this app for use in trying to find similar EDM songs through its musical elements and waveforms rather than just "important" factors like Key or Tempo.

Next steps.

Songs need to be downloaded for them to be embedded and searched, and Spotify's API actually does NOT let you download music for ML purposes.

While I had found a temporary "workaround" - Googling "spotify downloader" and attempting to use their APIs or using Playwright/Selenium to automate downloading songs - the owners of said API has added protection.

Key Takeaways

  1. Dockerized and deployed a live web app. DigitalOcean is so much cheaper than AWS.
  2. ML computation is more intensive on hardware than I thought --> Pre-compute or use smaller models for demo purposes.
  3. My definition of an API endpoints was misconstrued --> I'll work on this for the next project.

Contributors

mestvnvo

63 commits

mestvnvo/Bop-Dropper

Music Recommender System: Dropping some EDM bops near you!

2

stars

63

commits

Python

primary language

Jul 9, 2025

updated

www.bopdropper.com

README

Bop Dropper

Dropping some EDM bops near you!

Paste in any song from this playlist! https://open.spotify.com/playlist/2c8qTVOuNvJIl8UepuzHBb?si=8e68fd3da0b44dfe

Notes

Why?

I recently started DJing and created this app for use in trying to find similar EDM songs through its musical elements and waveforms rather than just "important" factors like Key or Tempo.

Next steps.

Songs need to be downloaded for them to be embedded and searched, and Spotify's API actually does NOT let you download music for ML purposes.

While I had found a temporary "workaround" - Googling "spotify downloader" and attempting to use their APIs or using Playwright/Selenium to automate downloading songs - the owners of said API has added protection.

Key Takeaways

  1. Dockerized and deployed a live web app. DigitalOcean is so much cheaper than AWS.
  2. ML computation is more intensive on hardware than I thought --> Pre-compute or use smaller models for demo purposes.
  3. My definition of an API endpoints was misconstrued --> I'll work on this for the next project.

Contributors

mestvnvo

63 commits

Languages

Python

60.4%

HTML

16.8%

CSS

11.0%

JavaScript

11.0%