Closed milasudril closed 3 years ago
If this happens it'll have to be done upstream at zita-at1. This is basically just a "LV2 shell" around the upstream project and I prefer to not fork the project, but rather add to it.
Then again re-tuning is in zita-at1 is basically just re-sampling and looping, so conceptually it's not feasible with the current approach.
Do you know anything about algorithms that can be used for formant preservation, or any tips on where I can read more? I have thought of implementing something on my own. I need a native version of KeroVee.
@timofonic merge what again?
If you mean zita-at1, there have not been any upstream changes since 2014: http://kokkinizita.linuxaudio.org/linuxaudio/downloads/index.html and there are no plans as far as I know to change the algorithm or shoehorn some format correction onto the resampler.
I don't understand your rant either. While it's true that some plugins do use unsuitable kitchen-sink frameworks, x42-plugins do not depend on GTK nor QT nor any framework. Just openGL, compatible back to 1.4 (early 2000s) and pango/cairo. If you want ready-to-run binaries: http://x42-plugins.com/x42/x42-autotune
If you don't like the GUI layout or design, complaining does not help, constructive criticism, mockups or suggestions how to improve visibility, learnability/discoverability control and interaction will go a lot further than just mentioning a the meaningless u* buzz-word.
@milasudril, for auto tune plugin with formant correction see TalentedHack. Unfortunately it have artifacts.
Also keep in mind that this auto-tune works with instruments that don't have formants or fixed resonances.
It would be really nice to add formant preservation. That is beneficial also for small corrections.