3.8 / 5

Harmony Engine

Play single notes, hear full harmony voices in real-time

Useful

Instant

The Problem

Hearing harmony requires arranging or multiple performers; parallel intervals sound amateur

Current Solutions (Not Great)

Harmonizer pedals (parallel only), arranging by hand (slow), practice with recordings

Who Needs This

Singers learning harmonies, composers sketching, choir directors, solo performers

You play a melody line; Harmony Engine adds harmony voices above, below, or both—not just parallel intervals, but intelligent voice leading that follows harmonic rules. Play a simple melody and hear it transformed into a Bach chorale, a barbershop quartet, a gospel choir, or lush jazz voicings. The AI understands harmony theory: it resolves tendency tones, avoids parallel fifths, creates contrary motion, and makes your single line sound like an ensemble. Perfect for vocalists practicing harmonies or composers hearing ideas fully arranged.

Honest Take

TC-Helicon and other hardware harmonizers exist, but they mostly do simple parallel intervals. The intelligent voice leading angle is unique, though you're entering a somewhat saturated vocal effects market. Browser-based is novel and could work for the education/practice crowd who won't spend $300+ on hardware.

Monetization Ideas

Ways to turn this into revenue

Freemium

Free basic, $5-15/mo for pro

One-Time Purchase

$9-49 per license

Subscription

$5-29/month or $49-199/year

Features

Key features that make this app valuable

Real-time harmony voice generation
Multiple harmony styles (classical, jazz, gospel, pop)
Intelligent voice leading
Selectable number of voices (2-6)
Above/below/both options
Key and scale awareness
Audio input support (sing and harmonize)
MIDI input for instruments

Build Prompt

Use this prompt with an AI assistant to start building

Build a React PWA called 'Harmony Engine' using WebInfer, Web MIDI API, and optionally MediaDevices. Input modes: MIDI keyboard or microphone (pitch detection via autocorrelation or ML). For each melody note, determine harmonic context. Use generateObject: { melodyNote: number, key: string, impliedChord: string, harmonyVoices: [{ pitch: number, voiceNumber: number, interval: string }], voiceLeading: { motion: 'contrary'|'similar'|'oblique', resolution?: string } }. Voice generation rules by style: Classical (strict part writing, avoid parallels), Jazz (extensions, altered tones), Gospel (thick voicings, chromatic movement), Pop (simple thirds/fifths), Barbershop (close position, ring). UI: visual representation of voices on staff or piano roll, style selector, voice count (2-6), register preference. Latency: 50ms buffer for simultaneous sound. Audio mode: pitch detection for vocalists, slight latency for processing. Output: MIDI or audio via Web Audio synthesizer. Educational mode: display interval names and voice leading analysis.

Score Breakdown

Pain

Clarity