The Rejection Remix

The Rejection Remix is built on a robust, script-driven architecture that combines custom generative music logic with cloud integrations. The technical stack includes:

• Generative Audio Engine (Tone.js): The primary audio synthesis is handled by Tone.js, a powerful Web Audio API framework. Our script creates multiple, synchronized polyphonic synth instances:

  1. PolySynth(Tone.Synth) for the complex, layered melody.

  2. MonoSynth for driving basslines.

  3. A custom 8-bit drum and percussion sampler, including multi-sampled congas.

• Scale and Rhythm Quantization: A deterministic logic system maps input text metrics to specific musical parameters. Our Scales API provides a selection of complex, non-standard scales (Blues, Bebop, Harmonic Minor, and exotic options like Whole Tone and Chromatic), ensuring distinct sonic textures for every generation. A customizable Rhythm Style Selector determines the pattern library (e.g., "Boss Battle," "Island Racers"), dynamically calculating phrasing and conga parts.

• Melodic Mapping: The core generator uses text input analysis. The system parses text into words and sentences, calculating metrics (sentence count, vowel density, punctuation types) to derive deterministic patterns.

  1. "Staccato" Standard Generation: A custom function determines Rhythm Profiles and Melodic Contours (Rising/Falling). Specific rules trigger deterministic melody lines: words starting with vowels generate quarter/eighth notes; rare consonant clusters trigger scalar and leaping patterns.

• The "Power-Up" Mode (Structural Parsing): Our advanced generation algorithm, the "Power-Up" System, creates four dynamic structural variations of a base pattern:

  1. 2nd Measure Legato: Elongating standard phrasing.

  2. Arpeggio: Generating ascending chord-tone patterns.

  3. Inverted Harmony (The R_COMPLEMENT): Creating melodic lines derived by inverting standard harmony relative to the progression root.

  4. Diatonic Sampling (The L_COMPLEMENT): Generating non-repeating scalar patterns based on unique vowel counts. These variations are sequenced into a complex structure (e.g., A-A-B-Comp_A-A) using a Cheat Code input system, allowing for deep structural customization.

• WAV and MIDI Export: After generation, the mix can be exported as high-fidelity WAV audio using the Tone.js Recorder function. For advanced composition, the WAV+ logic generates and exports separate, time-synchronized MIDI files for each layer (Melody, Bass, Drums, Percussion) on General MIDI Standard channels.

• Cloud Save (Google Drive + Sheets Integration): A complete request pipeline links the web app with a Google Apps Script backend.

  1. The music's unique settings are captured.

  2. A 15-second mix is recorded and encoded as Base64.

  3. The backend (via google.script.run) creates a dedicated folder in Google Drive, saves the WAV file, emails the link to both the user and Ryan Fox, and logs the request details to a specific Google Sheet (Date, Song Name, WAV Link, Email, Style), creating a fully automated request pipeline.