Music seem to be just another language. One of its building blocks are chords. So we might try to apply some computational linguistics tools upon music. One of such tool are n-grams.
It would be interesting to:
- extract the n-grams from chord sequences
- analyse their distribution
- how those distribution differ between songs or genres?
- could the n-gram distributions be used to predict next chords in sequence?
It seems there is some research in this area:
- Johan Pauwels, Jean-Pierre Martens, Marc Leman: Modeling musicological information as trigrams in a system for simultaneous chord and local key extraction