After 4 months of nights and weekends, I finally launched SoundScript
For the last four months, I’ve been building SoundScript alongside my full-time engineering job.
The idea started with a simple question:
What if music could be treated like source code?
Today’s AI music tools can generate interesting results, but they’re often non-deterministic. The same prompt can produce different outputs, making reproducibility and version control difficult.
I wanted to explore a different direction.
So I built SoundScript—a deterministic text-to-music language where the same source always produces the same output across supported platforms.
Some of the things I focused on:
Deterministic compilation
Cross-platform CLI (Windows, macOS, Linux)
Browser-based WASM playground
Extensible vocal engine
Reproducible outputs suitable for software workflows
I’m not claiming this is the future of music technology. It’s an experiment in applying software engineering principles—determinism, portability, and reproducibility—to music creation.
As a solo developer, balancing this project with a full-time job has been challenging, but incredibly rewarding. Every feature taught me something new about compiler design, audio systems, and building developer tools.
I’d genuinely appreciate feedback from developers, musicians, and anyone interested in programming languages or music technology.
What features or workflows would make a tool like this genuinely useful to you?
I like that you're applying software engineering principles to music instead of just adding AI generation.
Deterministic output is an interesting shift because it makes music reproducible, reviewable, and versionable—the kinds of properties developers already expect from code.