Back to audio today, and forseeing a blockage in programming I dedicated this Pact session to enabling future creativity. At the same time, I was able to assess the workflow involved in creating track audio components for the Python software.

Basically, at some point soon I'm going to need a full set of track audio files to build the full construction routine in Python. The creation of these files represents a challenge primarily of synchronisation: keeping all the audio aspects together.

For most music creators this will likely be less of an issue, but my specific studio setup makes it more of a problem. Because Prism Exhilarated was created in Native Instruments' Maschine using some external MIDI sound sources, and I pump my audio interface outputs through a summing mixer, an extra step or two must be added. Here's the workflow:

1. Bounce the track backing and all unique process option parts out of Maschine through my summing mixer (I use Audio Hijack to capture these bounces).
2. Import all those files into a DAW (in this case Presonus Studio One) and resync them.
3. Ascertain the amount of gain to apply to all components to achieve 0.0dB output of the track as a whole, and apply to all components.
4. Export the parts. Make sure a unique process's audio files all start at the same time.
5. Note the start time of the unique processes, to put into the Python software's config file.

So today I ran through this workflow with the third track from Prism Exhilarated: 'Xplorer'. Now, bouncing all that audio takes time (see all the files created below), so I could only manage to get through steps 1 and 2 (as you can see from the image above).

I also came across a potential issue: the unique process in Xplorer has in it a compressor sidechained from another channel. That sidechain processing wasn't applied during the bounce, because the track supplying the sidechain signal was muted. So I may have to reapply that somehow in Studio One.

Hopefully in the next session I can complete the workflow steps.