The Neural Diva Sings

The Neural Diva Sings

What you are about to hear is: "Where are the great ones The fire-minded ones"

...

It is difficult to estimate in human terms, the amount of effort required to make my Neural Diva sing.

AI and I have to do a solid dollop of refactoring, as we chase a feature or too.

The AI rewrites code well, humans not so much, we are just too slow.


You can’t have just one singing voice in an Opera, and it is a tragedy, when we are forced to rely on a single orchestra.

The world needs movements that move, that explain where to go in life that does not result in a loop.

That can’t be presented in a singe format, there must be gently robotic voices here.

Pop, Folk music, Techo of course, UFO sounds all the way, and then the pure voice of a Cathedral Neural Diva.


I will let AI coin the term blur, here.

Depending on what story in being told... a Diva may need to blur her words.

This does not mask the tragedies that she sings about, it is just that every word here becomes unnecessary.


I am concerned about the musical side, and I am working on a second program, a proof of concept.

It is a classical music generator, a classical response to the Drums, Bass, Chords and Melody.

AI made it clear, that this is not as simple as it sounds, because these can’t be any Drums, Bass, Chords and Melody.

They each need to be manufactured for each other, so if we didn’t have AI this would be the impossible.

With AI, generating or even composing compatible music parts, is just the next step, that only takes a moment.

In Classical music we don’t actually have Drums, Bass, Chords and Melody.

We have Pulse/figure, Bass voice, inner voices, and theme, in Opera, we have the orchestral pulse, harmonic floor, Orchestra color and vocal line.

You understand now, why this is difficult to estimate, we are composing on the scale of multiple lifetimes.

My proof of concept resulted in valid music, it works beautifully.

I hear it as UFO sounds, the root electronic sounds, that eventually become the song, or midi if you ever played them.

And the vocal line is the perfect launchpad for the diva, the pulse flor and color remain stable and she takes off fully compatible.

And compatibility here does not end at opera and classical, I will instruct the AI to cross the genre line.

Allow me now, to tell you what four pieces in this model, make up other genres that I hope to integrate with.

EDM and pop is more correcty, Drums, Bass, Chords and Lead.

Rock has, Drums, Bass Guitar, Rhythm Guitar, and Lead/Vocal.

Country has, kick/snare/clap, Bass, Guitar/Piano, and Vocal Hook, or fiddle.

And Jazz has Ride/groove, Walking Bass, Comping chords, and Solo Line.

In my little four piece theory, in teory, AI is able to bridge the genres.

So that I can swap the Opera Orchestra Pulse for Rock Drums, and harmonic floor classical music cello bass line.


And of course, tha vocals will gravitate between Diva and Solo Pop, based on what need to be said.

And what needs to be said, is that schools do not work, and what follows is the slow unraveling of people who are meant for greatness.

This is nothing new, it is as old as the murder Socrates.