SukhmKang / music-analysis

Project that detects beats within an inputted song and finds its tempo (beats per minute). This repository accompanies our paper, "Global Tempo Estimation Algorithm for Popular Music with Noise-Filtration and Spectral Clustering Algorithm."

music-analysis

Ishan Balakrishnan & Sukhm Kang

Project that detects beats within an inputted song (.wav file) and finds its tempo (beats per minute). Annotations for the testing datasets are available in a parseable format at this link.

Built With

Python (https://www.python.org/)
Pandas (https://pandas.pydata.org/)
Wave (https://docs.python.org/3/library/wave.html)
SciPy (https://scipy.org/)
MatPlotLib (https://matplotlib.org/)

Algorithm Summary

Please see our academic paper, "Global Tempo Estimation Algorithm for Popular Music with Noise-Filtration and Spectral Clustering Algorithm," for an in-depth summary of the implementation, methodology, tuning, and testing of the algorithm.

Results

Dataset	Type (Training/Testing)	Algorithm Accuracy
SPOTIFY TOP 100 2016	TRAINING	87/100
SPOTIFY TOP 100 2018	TRAINING	93/100
BILLBOARD HOT 100 YEAR-END 2021	TESTING	91/100
BILLBOARD HOT 100 APRIL 2022	TESTING	89/100
SPOTIFY '00s ROCK ANTHEMS	TESTING	85/100
GIANT STEPS	TESTING	562/662

Authors & Contact Information

This project was authored by Sukhm Kang and Ishan Balakrishnan.

Sukhm Kang
Mathematics @ The University of Chicago
https://www.linkedin.com/in/sukhm-kang

Ishan Balakrishnan
Computer Science & Business @ University of California, Berkeley
https://www.linkedin.com/in/ishanbalakrishnan

Feel free to reach out to either one of us by email @ ishan.balakrishnan(at)berkeley.edu or sukhmkang(at)uchicago.edu!