September 22nds

 Noondreams (9-22-1995) by meta4s

9/22/1995 

For the Fall Equinox:

9/22/2017

“Soulful” or any other words or concepts attached to human emotions can possibly be built into machines, but it begs the question why. I have been attempting to compose piano pieces with algorithms and I still can’t answer the Whys. I prefer playing them myself because Soul is built in. 

[9/22/2025: The latest foray into AI-generated songwriting is an experiment with Esperanto translations of some of my 90s lyrics. If you were to do this manually 30 years ago you'd have to find someone to translate them, then write melodies. Since I don't speak the language (does anyone?), it would be impossible to sit down with an instrument and sing against it. AI can do that, but it's a soul-less mechanical process--something I sometimes find hollow and creepy. The music itself works as a kind of addicting confection which hides the fact that the sugar isn't good for you. The translations are also clumsy and too wordy. "Kapitulacu al la forto". Try spitting that out while playing guitar, but AI can do anything. The sound of the words are interesting--at once Spanish, Italian, and Greek. If the language had a place it would be the Roman Empire around the Mediterranean. 

The whole process of AI music generation is production. You are the producer, which requires wearing different hats. Choosing one version over another is one of production, which is all this is. It isn't that you're shaping the sound in a recording studio. It's the perfect example of using the internet as an instrument.] 

9/22/2025

What's the difference between music created with AI, where the music arises from the lyrics that are provided, or generated, and the natural rhythms in words that I use? If I use my words, which create my rhythms. Whether you write music manually or you generate it with AI, the rhythms in the words are still driving the music. So if you take away the words, what's the difference in the quality of the music between the two? I would say that the music that is generated from my words with rhythms is better than the music that's generated with AI. Generally speaking, the musical quality, regardless of the quality of the recording, is less interesting with AI. When I listen to just the instrumentals in the Esperanto music, it is very generic, even though the language has its own unique rhythms. People could take that music and easily map new lyrics over it, whereas it would be more difficult for someone to assign new lyrics to my original music because it’s not generic.