A (brief?) return to music composition

One of my main goals is just to be the shoulders on which giants stand. Everything I discover about music, however small, will go on to inspire generation after generation to come, sending waves and waves of inspiration throughout the rest of history.

I learned something else. The numbers representing the chord transformations cannot go into the negatives, because doing so has a harsh effect. Here's the improved version of that song generated from my simple input melody:

A three-voice canon I composed in a more traditional fashion:

Well, what's an invention fugue?😅 I only know Inventions OR Fugues but never know they can be combined...

On 6/28/2025 at 9:41 PM, Henry Ng Tsz Kiu said:

Well, what's an invention fugue?😅 I only know Inventions OR Fugues but never know they can be combined...

 

An invention fugue is a musical invention of mine created while I was undergoing a fugue; hence the name. (It might have better been called a fugue invention, but oh well.)

At any rate, I've gone back to computer programming to study and create music. My latest creation is, in its core idea, a most unique and interesting one if I do say so myself. The main idea behind it is that melody and harmony are identical to each other except they span different axes. No, I don't mean that a melody is an arpeggiated harmony. If it were, melodies generally wouldn't produce such dissonant and incoherent harmonic progressions when the notes comprising melodies are superimposed. No, the way to turn a harmony into melody or a melody into a harmony is much more complicated and esoteric than that. There is a certain principle whereby one can translate one into the other--and even superimpose the two translations over each other to create a coherent harmonic progression rather than a mere static harmony or a monophonic pitch progression. I have--and I think I am the only one who has done it--discovered the principle that makes doing this possible. Using this principle, I have designed a computer program in which I input harmonies, turning them into melodies and melodies, turning them into harmonies, and in which these two translations are played simultaneously to create counterpoint. The amount of effort involved in using it is very little. All I have to do is provide a small amount of detail, and the program fills out the rest.

The piece below is the first remotely serious one I used the program to write. It's very short, and, yes, the instrument that renders it is a harsh-sounding one due to it being the best default instrument. But I think it has some merit to it at least as an illustration of my idea. It should be noted that the piece involves self-imitation, which the principles behind my program, and the program itself, makes pretty easy to implement. There is a motif used once by one voice, slowly, and, at the same time, five times at different pitch heights by another voice. It also has exactly four voices, two of which required a small amount of design on my part (much less than would ordinarily be the case when writing for a pair of voices), the other two of which were generated completely automatically.

I discovered that the base pitch needs to be ideally calibrated for maximally good effect. Here's the the same piece with ideal pitch collaboration. It sounds way better, way smoother and far less dissonant:

I've made further progress, having discovered a crucial principle behind music composition. The following piece, though very short, is composed with six voices and includes a great deal of so-called xenharmonicity. Better pieces will no doubt follow.

Three samples:

The forum refuses to display this one, so here's a link:

https://soundcloud.com/user-321964225/experiment-a