Six-Phase Rhetorical Cue Planner for Production Music

Concept

A structure tool for short-form music, working in two modes.

Plan mode. Takes a brief (length, use case) and generates a six-section arrangement map with suggested timings and measurable energy targets. Sections are named in plain English — Opening/Hook, Unfolding, Theme Statement, Reinforcement, Contrast/Tension, Climax & Resolution — with the classical Latin terms (exordium, narratio, propositio, confirmatio, confutatio, peroratio) available as an optional shared vocabulary for revision notes.

Lint mode. Reads a draft arrangement (MIDI or stems) together with the plan it was written against, or with section markers the composer places. It does not attempt to hear 'contrast' or 'climax' as subjective qualities. Instead it computes deterministic proxies per section — RMS loudness, instrumentation count (from stem tracks or MIDI programs), tempo, and harmonic density, plus spectral novelty for boundary detection — and checks conformance: does each section hit its declared energy target? Is at least one section measurably different from its neighbors (the contrast proxy)? Does the final section carry the loudest peak (the climax proxy)? Every flag cites its measurement, e.g. 'sections 3 and 4 differ by under 1 dB and use the same instrument set: possible missing contrast,' so the composer, not the tool, makes the final call.

Keeping false positives survivable. Thresholds are configurable per library; composers can annotate a section ('intentional quiet contrast') to suppress a flag with a reason; and catalog-scale runs report aggregate patterns rather than hard pass/fail verdicts. The tool is advisory: it surfaces data, humans keep the judgment.

Honest limits. The proxies are crude. A quiet-but-tense contrast (dropping to a solo instrument over dissonance) may not register as an energy change, which is why the tool also watches instrumentation and harmonic-density shifts — but contrasts achieved purely through register, texture, or voicing can still be missed. This bounds the tool's claims: it can verify that a draft matches a structural plan on measurable axes; it cannot certify that a cue 'feels' contrasted or climactic.

Why now

The treatise [0] documents rhetorical thinking applied to melodic works, including the arrangement phase, as a sequence of rhetorical processing phases — evidence that structure-level rhetorical organization is a real, historical approach to instrumental music, and the inspiration for organizing a cue around rhetorical functions. On provenance, to be clear: the available abstracts do not enumerate a specific six-part Latin scheme, so the six-section grammar above is this tool's own adaptation of the traditional six-part speech outline (opening, background, thesis, support, counterargument, close) — a design choice, not a structure taken from the treatise. Dewey's analysis of the formal structure and characteristic effects of the arts, including music [3], supports the premise that arrangement-level structure, not just melodic content, shapes the listener's experience. And because the checker is deterministic — no training data, no model — it suits production libraries that must QA thousands of cues cheaply, giving editors a measurable checklist instead of taste arguments, bounded to what can actually be measured.

Open questions this iteration researched

• How well do loudness/instrumentation/tempo proxies track what listeners and librarians actually experience as 'contrast' or 'climax'? (Needs validation against human judgments; until then the tool's claims stay bounded to plan-conformance.)
• Does the production music industry prioritize structural consistency over composers' creative intuition in high-volume libraries?
• Is there a standardized MIDI or stem metadata format for section boundaries, or must the tool rely on its own plan files and markers?
• Would a deterministic checker still produce too many false positives at default thresholds, leading composers to ignore it? (The mitigations above are a design response, not a validated answer.)
• What is the economic value of reducing taste-based revision cycles in the B2B music licensing workflow?

No new supporting or conflicting evidence was found this iteration; these questions remain open.