Chasing feelings with AI

Ira Glass talks about the gap between taste and ability that haunts creative work. You know what good sounds like, but your skills disappoint you. For years, this meant choosing: either spend time mastering the craft, or live with the frustration of unrealized ideas. I've been wondering if AI might change this equation. Maybe you can translate a vague feeling into something functional, at least in domains where you know what you want but lack the tools to make it.
"All of us who do creative work, we get into it because we have good taste. But there is this gap. For the first couple years you make stuff, it's just not that good [...] But your taste, the thing that got you into the game, is still killer. And your taste is why your work disappoints you. [...] Most people I know who do interesting, creative work went through years of this..."
Ira Glass
I've been playing guitar for years. There's something satisfying about working through chord progressions, understanding why certain combinations create tension and release. But I've lived in Ira Glass's gap more often than I'd like to admit.
The ideas in my head would outpace my technical ability to express them. I'd sketch out a riff that felt promising, then open my DAW to expand on it. That's when I'd hit the wall. I could hear what the track needed: a drum sound that felt "saturated" and "dark" and "encompasses a room," but I had no idea how to actually create that texture.
So I'd start clicking through drum samples. Dozens of them. My ear would get fatigued after twenty minutes of auditioning kicks and snares, each one sounding more generic than the last. By the time I'd exhausted the sample library, I'd lost all connection to the original riff idea. The creative spark that brought me to the DAW was gone, replaced by browsing presets.
This is usually where I'd close the laptop. My intuition was pulling me toward something but I had no path to follow it. For me, the gap between inspiration and execution often killed the joy before I could even explore the idea.

From feeling to sound

The demo below shows how AI translated those descriptive words into working audio code. Click through the examples to hear how each iteration got closer to the feeling I described.
Kick Drum Controls
ATTACK
1ms
DECAY
0.8s
SUSTAIN
0.2%
RELEASE
0.1s
FREQ
60hz
BEND DEPTH
2.5x
BEND RATE
0.0s
CLICK
0.4%
RESONANCE
0.8
Effects
ECHO TIME
95ms
ECHO MIX
18%
step sequencer
bpm
79
For me, this opened up a different way of working. Instead of losing creative momentum to sample browsing and technical barriers, I could prototype feelings directly. The spark that brought me to the DAW stayed alive. I could chase an idea while it was still vivid rather than losing it to searching for the right sound.
Maybe AI offers a different path through Ira Glass's gap. You start with the feeling and let AI handle the technical translation, then use that understanding to guide your traditional practice with clearer intention.
The gap Glass describes is still there and true mastery still requires discipline, but now when your intuition pulls you toward something, there might be a way to follow it immediately. At least for me, the hobby feels more joyful when I can reach beyond my current abilities without getting stuck.
This article was written based on a conversation with Bojan Joncic