growing up, i learned to sing Carnatic
(Indian classical) music. Carnatic music songs are often based on a scale, or a raga
. naturally, i wondered whether other songs i listened to adhered to carnatic music ragas.
raga detector started as a project for my jazz class--i built a website using Flask and the Spotify API in order to determine the closest melakarta
raga to a song.
as a part of my final project for 6.8371 (computational photography), i wanted to use the techniques we learned for image manipulation in order to process music.
using the short-time Fourier transform, i wrote a program that will take a song's wav file as an input and generate its spectrogram
when i was a part of the hackmit organizing team, i wanted a way for teams to be individually matched with a mentor who would help them throughout the course of the hackathon.
i helped build an internal app using Flask and PostgreSQL for the team to use that would automatically pair teams and mentors based on their individual preferences/restrictions.
using tensorflow, i wrote an app that uses the timbre vectors generated by the Spotify API for a song in order to detect what the most dominant instrument during that segment of a song is.
look to the left to see an example of what YOU could see if you used it too!
i created a command-line urban dictionary in an effort to play around with web-scraping and python, although i'm not sure why anyone would use it : D