The authors evaluate a frozen Explainable Melody Harmonizer (EMH) on 19 HookTheory excerpts containing aligned melody and chord annotations. EMH generates chord sequences from melody alone using explicit criteria for melodic compatibility, harmonic function, persistence, cadence, meter-aware segmentation, and Viterbi-like optimization. Human chords are withheld during generation and used only for post-hoc comparison, yielding 439 evaluable harmonic segments.

Exact chord agreement reached 20.50%, chord-root agreement 31.66%, and harmonic-function agreement 47.62%, with functions assessed relative to the key inferred by EMH. Decision-level analysis found 203 segments (46.24%) where the human chord lay outside the model's candidate vocabulary and 146 (33.26%) where it was available but not selected. This distinction separates representational limitations from differences in chord selection, without treating the human reference as the only musically valid solution.

The work shows that a model developed for individual songs can be applied to independently collected human annotations without changing its rules. Its main contribution is a reproducible, interpretable framework for corpus-based evaluation of harmonic decision-making, offering a clearer view of where such models fail and why.