This won’t be the most interesting of updates, but I’ve been working on the radio drama content creation pipeline, which results in what you see below. Obviously, this is a work in progress, meant to be thrown away.
More important than quality writing was a realistic test of every part of the pipeline, from script writing, recording, machine learning voice replacement, audacity editing, drop accumulation, programmatic markup, and finally in-game playback. This experiment was enough to show me a few flaws in my original idea.
The basic premise is that two podcasters/radio jockeys are trying to tell the story of a girl’s escape from Epstein Island, but keep getting interrupted by an ancient demonic force in the shape of a homeless druggie, and the feds outside who are preparing to Waco them. Sort of like this, but more complicated.
The original plan called for the demon to take control of their sound board, which would have contributed to the humour, as well as helped me cover up issues with the audio. I’ll probably scrap that idea, for a few reasons. First, I simply do not have any appreciable library of sound drops to utilize. Second, it’s more difficult than I thought to integrate that into the scenes. Third, the sheer amount of editing required to make these jokes land might be a bit too much.
My thoughts are still a little vague on this front, so more testing is required. I’ll take a step back, do another pass through the enemy AI/design, and re-assess.
