vl8 VS rec

Here are a few examples: https://github.com/rec/vl8/blob/main/studies/study1.py

https://github.com/rec/vl8 - the advertisements about what it "can" do are "true", but you can't do any of these things from the command line, it's not at all release worthy.

Here is link number 1 - Previous text "vl8"

I have a alpha-quality system myself for processing audio in numpy buffers called vl8 (getit?).

I'm hoping to fit my newest working project, https://github.com/rec/vl8, into this framework.

Me, I'm a long-time developer with a lot of little production Python libraries and an alpha level audio DSP CLI called VL8, which seems to have almost the same API as PedalBoard's, so I'm thinking of porting it to use that.

I've been writing C++ since the 1980s. I wrote a commercial application with Juce: https://github.com/rec/rec

It's greatly improved workflow. But not "just". numpy is a better tool for processing arrays of numbers than anything C++ has, particularly if you aren't an expert. (And, wait, I am an expert C++ programmer, and I still prefer Python! Aside from being employed by others, I also wrote a moderately-sized, marginally successful commercial digital audio application in C++.)

Otherwise, you're really getting into C++ world. Probably the best solution is the JUCE system - I wrote a whole application for this some years ago, but it's writing VSTs or AUs is much easier, though still tricky.

It's here. I made some decisions I don't love in hindsight as a solo developer but overall I'm happy with the code.