Hooktheory Uses Data to Quantify What Makes Songs ‘Great’

If you listen to the Top 40 enough, you might start hearing some similarities – and if Hooktheory has anything to say about it, you’re right on the money. The Berkeley startup, which specializes in building software and providing learning materials for songwriters and musicians, believes that the things that make music popular can be quantified using hard data. To prove this, the company has developed a crowdsourced database of “Theorytabs,” touting it as the world’s largest community-sourced database of popular song analyses… and planning to enhance it using machine learning.

The Hooktheory analysis database began as a “labor of love” by Hooktheory co-founders Dave Carlton, Chris Anderson and Ryan Miyakawa, based on the idea that “conventional tabs and sheet music are great for showing you how to play a song, but they’re not ideal for understanding how everything fits together.” Over time, the project snowballed into a community effort that compiled tens of thousands of Theorytabs, which Hooktheory describes as “similar to a guitar tab but powered by a simple yet powerful notation that stores the chord and melody information relative to the song’s key.” 

When viewing Theorytabs – which cover artists from Taylor Swift and Queen to the composers for Game of Thrones and The Legend of Zelda – users can “zoom” in or out, loop the tabs, change the keys, adjust the tempos, tweak mixers, and more. They can even choose to listen to simple piano versions of the Theorytabs or watch the Theorytabs synchronized with the songs’ music videos on Youtube.

An example of a Theorytab.

“The Hooktheory database owns more than [20,000] well-formatted tabs for popular music, which is a rare and precious data source for [many] tasks,” said Junyan Jiang, a researcher in computer music at Carnegie Mellon University. “The value of the Hooktheory database is unlimited.”

Now, the Hooktheory team is working with the machine learning departments at Carnegie Mellon University and New York University to take Theorytabs a step further. With help from data scientists, Hooktheory is aiming to leverage its massive database of annotated music to answer a central question: what, exactly, makes a great song?

“We’ve had an enormous response from our community of users who contribute to and maintain the Theorytab database every day,” says Chris, “We’re collaborating with researchers and developers in both the music and AI spaces, and believe that this data can reveal musical aspects that make us so drawn to our favorite songs, as well as provide a powerful tool for helping songwriters and musicians in their own musical pursuits.”