Song Recommendation Using Fourier
summary
Uses audio signals' frequency content to find and cluster similar songs without Spotify's metadata. It converts time-series audio into frequency-domain representations (via Fourier) and applies clustering with different distance metrics to group tracks by similarity.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 0:47Song organization problem
- 3:17Fourier transform explained
- 6:30Applying Fourier to songs
- 8:19Testing with music genres
- 10:26Library use and results
- 11:56Example with five genres
- 12:18Q&A
related talks
36mExtracting Relevant Metrics with Spectral Clustering
04mMusic, the window into your soul
14mThe nuts and bolts of music source separation
31mThe Fourier Piggies and Machine Learning
40mBuilding A Folk Music Recommendation System with LLMs
38mThe Magic of Music Matching
38mmusic2vec: A smart music recommendation and discovery
33mEvolution of Music, Cantometrics