Skip to content

Music recommendation system and visualization web app

License

Notifications You must be signed in to change notification settings

peter-ai/acoustic-print

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

40 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

acoustic-print

Acoustic Print is a data visualization web app that takes a unique approach to a traditional class of techniques in acoustic fingerprinting. Based on audio features of a song, such as energy, tempo, and valence, to name a few, a two-part acoustic fingerprint is constructed out of polar curves and visualized in 3-dimensions. Acoustic Print allows users to visually perceive similarities and differences between songs based on underlying characteristics.

Acoustic Print was developed in Python with the front-end built on Streamlit, using a MySQL backend.

Data on which this project is built is from the Free Music Archive (FMA), in particular, from the study FMA: A Dataset For Music Analysis by Michaël Defferrard, Kirell Benzi, Pierre Vandergheynst, and Xavier Bresson (github). Unfortunately, given the nature of the archive, you will likely not see many songs from the most popular performers and your favorite artists. A live API would have made this even more interesting. Due to rate limiting on the Spotify API and a policy against the use of Spotify data for machine learning, using live or recent data for this project was less feasible. Nonetheless, I hope you enjoy.

About

Music recommendation system and visualization web app

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published