just how on earth do u guys do it? building audio software? gods ya'll are

[aksh-sush]

I'm sorry if it's not appropriate for me to ask this here, ..but I found no answer anywhere else, so this is my very desperate attempt

I'm obsessed with any audio releated dev, the thing is as there's an abundance in resource of how to use python n make midi out of it and all that, I usually start in Python, pick a library, or find a GitHub repo that claims to do what I need in under a minute. But then it doesn’t fit my use case, so I end up writing scratch scripts just to get basic note detection working. I get lost in the process, and the script only works for the sample audio I used, because I didn’t account for all the variables in real songs. Eventually, I just stop

I want to do simple things like converting pitch maps from uploaded audio (not sine waves, but real audio) to possibly splitting it into constituents, as DAWs do, and then MIDI-ing it, providing MIDI and possibly pitches of a song split, all basic stuff that, when sequenced and well-built, makes software like FluidSynth itself. The more I look, the more lost I am..I don't want to build an application; I just want to learn the underlying methods and functions, such as breaking down a song into its notes. (like sheet music, but just the pitches will do)...I'm not sure... where did you start? How can I go about this? And also, a very dumb question, if I did maybe use liberosa and fft to do pitch mapping, where can I find more people like me....at least then I'd know what is realistic requirements are to be eligible to build stuff like these... like the learning curve is like super steep and how do u guys do it? Where do u begin? Where do u work? Where do ppl like you all socialize? (I know rants r for Reddit, but I could badly use your help.)