Song
♪
Loading…
Select a track to begin
0:00
0:00
| marks the best parts
Some of my feelings arrive at full strength. No warning, no volume control. The overwhelm that freezes me before I even start. The moment at the ocean when nature is so much bigger than anything in my head. The late ride home after a good evening with people, tired but still full of energy, the city quiet and my mind running free.
This album is about those moments. Not the managed versions of them. The real ones.
Three tracks, each one a different state I know well. Too Many Thoughts is the freeze, the noise inside, too much at once and nowhere to go. Open Water is the opposite, out in the elements, body fully awake, nothing between me and the world. Night Run is the warm electric feeling of riding home after a good night out, tired and happy and somehow more alive than all day.
Together they are a map of how intensity shows up in my life. Not always comfortable. Always real.
The hard part was not the prompting. The hard part was knowing what I actually wanted to feel, and then staying with that across three tracks instead of chasing whatever sounded good in the moment.
The biggest thing I learned is that being specific matters more than being detailed. Vague mood words did almost nothing. What worked was concrete physical language: the actual drums, the BPM, the texture. The tools have clear personalities and knowing which to reach for made a real difference. Chaining Suno and Stable Audio together for Open Water was the most interesting moment of the project. Suno has a very recognizable sound to it. Stable Audio produces something higher quality and less obviously AI. Running the Suno track through Stable Audio gave the result a depth and weight the Suno version alone did not have.
Originally I wanted to stay mostly instrumental. Working with lyrics in Suno was an interesting part of the process and came together faster than expected for Too Many Thoughts. In Open Water the voice works completely differently, no words, just vocal texture woven into the sound. One track uses voice to tell a story. The other uses it as pure feeling. That contrast ended up being one of the more interesting things about the album.
I tried more tools along the way. ElevenLabs in the first assignment, Audiostellar, Holly+. But every additional tool means more time and more complexity for results that are not always better. Holly+ did not fit my album concept at all. In the end I decided to focus on Suno and Stable Audio and do those two well rather than spread across everything.
What I am missing from the course is the depth to actually go further. Without a professional music background it is hard to know how to edit and shape tracks beyond the prompt. That knowledge gap became very visible here. Without Claude to guide the prompt structure across different tools this assignment would have been very difficult to execute.
Directing AI music is mostly about knowing what you want before you start. The tools follow. The intention has to come first.