CloudHolic/harmonyrl-chromatic
Reinforcement Learning • 18.4M • Updated • 41
Melody harmonization models: chord progressions for a melody, in 3 styles.
Note Rich reharmonization with many chords from outside the key (RL fine-tuned).
Note Consonant progressions that resolve (RL with a per-note reward after RL-Chord).
Note Close to the training data (supervised only).