Chilluminati91
Chilluminati91
Can be solved by following [this post](https://github.com/AUTOMATIC1111/stable-diffusion-webui/discussions/9605#discussioncomment-5688252).
Start with 35 then work your way down until you dont get memory allocation errors anymore. Then substract 2 and you got your value.
> try pip install httpx==0.23.0 Works, thanks! Should probably be added to requirements.txt
Will take a look at it, but to keep clutter to a minimum I would instead introduce a single button in the settings tab that saves all current values to...
Shouldnt this be pretty simple in general? Calculate a mean f0 from the training data. Then calculate mean f0 from your clip before inference and shift by the difference.
Not too sure whether its needed. Would require one config for each basenote + scale (major, minor, natural minor, harmonic minor, mixo-lydian, etc.) or an intelligent algorithm to calculate them....
It kinda depends on the skill of the original singer. If you dont have perfect pitch but want to get a slight autotune you can set the slider to 0.3...