I want to clarify what I’m trying to accomplish, and what I’m specifically not trying to accomplish. Commercial versions of this technology do a sophisticated analysis of the sounds they detect, so that they can do things like distinguish between fireworks and gunshots using shock wave profiles and such.
At least for this first iteration, I’m not looking to do anything nearly that complex. I just want to be able to pinpoint any loud sound in a reasonable range. Future iterations may do something as subtle as attempting to disentangle overlapping sounds, or deal with long continuous sounds (like a running motor, jackhammer, etc.), but those are stretch goals.