// (this + some of my citations was what I submitted for the actual mit portfolio form)
Goal
Create novel musical sequences given a seeded random number generator and a set of configurable options.
Restraints
My original technical portfolio submission was going to be a sight-reading trainer. However, I realized too late in the process that it was not possible for me to make it feature-complete and thus dropped it. On receiving a deferral, I learned that I could submit a project for RD consideration, leaving me with 20 days to complete a substitute project while still busy with midterms and other applications.
Timeline
I began work on the project on December 17th, 2025, and completed it January 4th, 2026. It took me until 6:16 on January 5th to make the frontend Apple-device compatible after realizing that Safari could not properly resize SVG files. I spent a total of 47 hours and 41 minutes over 13 days.
Methodology
Generate a sequence of n notes whose pitch is determined by weighted RNG (prioritizing changes in pitch that more often occur in actual music)
For each measure, attempt to find a diatonic chord (the diatonic chords available to use are based off the melody’s scale) that includes all the pitches in the measure. If impossible, attempt to find a chord that includes all the first (n) - (# of failures) notes.
Take each chosen chord and based off the 2nd octave, apply a random voicing in the range of 0 to (# of notes) - (1) (inclusive)
Apply transformations to the melody and chord progression to fit them in their respective series of measures
Merge either series of measures
Present product to front-end
