Before I say anything, most of the credit for this here goes to @fastfinge and friends and their Eloquence64 project, which until today I considered the most complete, stable Eloquence. However, I had a bug that only I seemed to exhibit regarding unicode handling, and Sam and co weren't comfortable with my fix. understandable.
@nick I really do want to get to the bottom of this. The main issue with your PR is that it does seem to introduce a click sometimes; if it had no side-effects, I'd just merge it. But between not being able to find anyone else with the issue, and the clicks, I need to hold off.
@fastfinge I figured it would be wasteful to not put this somewhere, so I figured proper attribution and put it out there. Also, I confessed that the more I dug in, the more I was concerned about just contaminating your code with an AI guessing, so I figured I’d let it rewrite the audio handling and host process so it would know what its own work look like.
@nick Does the fork with the upsampler have the same bug? I’m just curious because that guy rewrote audio processing as well. Honestly I’m probably being a bit too precious about not merging code. But I want the code in my repo to be as simple as possible. Because every time NVDA changes how addons work it has to be rewritten. I’m fine with my repo being the stable base everyone can build more feature rich forks on top of.
@fastfinge Said in the PR but yes, the upsampler fork does exhibit same bug. The bug came from an indexing issue, but fixing it, for reasons, generated clicking worse than the pR. I threw up my hands because I wasn't trusting my ears, but then dug in again when it looked like you, understandably, appeared as if you didn't want to dig into a PR for an issue you couldn't even test for.
@fastfinge@Nick And this is why I was sad when I couldn't find an SJAMS anywhere. That was one of the slickest interfaces to speech with mIRC. I felt your code wasn't pulling too many tricks, just being clever with what mIRC offered.
