TOPIC: How does Polyphone make low-def audio sound good?

I know it's everything to do with the sound engine, but like, the ability to make 12k Hz audio sound that good in other programs would be grand. FL Studio's DirectWave does something similar. I think it's maybe due to interpolation.

In other words, assuming it's possible, what sound settings would I use in, say, Audacity, if I wanted to get that same quality (without having to resort to recording my speaker output)?

Outch you just put your finger in a complex subject which is indeed the interpolation, which is used for

• resampling data for matching the audio output sample rate,
• changing the pitch (coarse / fine tune, modulations...).

At first I simply used a linear interpolation
en.wikipedia.org/wiki/Linear_interpolation

But in doing this, artifacts appeared when increasing the pitch of already high-pitched samples because of aliasing:
en.wikipedia.org/wiki/Aliasing

I noticed that Fluidsynth doesn't have this drawback so I analysed their code and Polyphone is now using what appears to be a 7-order sinc interpolation formula but don't ask me to explain how it is working!!
en.wikipedia.org/wiki/Whittaker%E2%80%93...nterpolation_formula

There are many interpolation functions and this one seems to respect the harmonic content of the sound wave particularly well.