CHORDONOMICON: A Dataset of 666,000 Songs and their Chord Progressions · HackerPapers discussions

CHORDONOMICON: A Dataset of 666,000 Songs and their Chord Progressions — mentioned by mkl

Fri, 18 Apr 2025 04:53:03 GMT

Yes, I was disappointed. The original paper https://arxiv.org/abs/2410.22046 did look at chord progressions. They also trained a machine learning model to predict the next chord. Some of the chord progression data is in graph form at https://github.com/spyroskantarelis/chordonomicon. The raw chord data is at https://huggingface.co/datasets/ailsntua/Chordonomicon/tree/.... It consists of one row per song containing a list of chord names in song order (no timing information) and Spotify ids for track and artist. It seems like Spotify has a different id for every released version, so it's really…

Paper and discussion history. A mention is not an endorsement.

CHORDONOMICON: A Dataset of 666,000 Songs and their Chord Progressions — mentioned by memset

Fri, 18 Apr 2025 02:54:50 GMT

The way this analysis, and the original dataset were created, makes no sense. This is, in part, not the author's fault, since the original data [1, 2] is flawed. First, the original data was constructed like this: "...The next step was to format the raw HTML files into the full chord progression of each song, collapsing repeating identical chords into a single chord (’A G G A’ became ’A G A’)..." Already this makes no sense - the fact that a chord is repeated isn't some sort of typo (though maybe it is on UltimateGuitar). For example, a blues might have a progression C7 F7 C7 C7 - the fact…

Paper and discussion history. A mention is not an endorsement.