Evaluate resamplers
Opened this issue · 0 comments
jdisanti commented
The current implementation is using Rubato's SincFixedIn asynchronous resampler with linear interpolation. There are a few other options, such as FftFixedIn, and FastFixedIn. The differences between them seem to be the algorithm, and whether or not you can change the frequency ratio after creating the resampler (which is a feature that isn't needed here).
I'm not sure if SincFixedIn is the best choice for this use case or not. Criteria to scrutinize are efficiency and output quality.