Feminine voices produce not really matching results
So I say your program and I say it works quite well so far, really good. So I made a plugin for the program called blockbench to use the output files from this program into keyframes of the model im using it for. Anyway to the point, I notice that voices of the higher pitched and feminatialy tend to produce incorrect mouth movements to what I am making. I might be dumb and not knowing the solution but just in case would you mind telling me?
I'm afraid this is a problem I'm aware of. The speech recognition engine used by Rhubarb works best for low-pitched male voices and gets less precise for female or child voices. I'm actively working on Rhubarb v2, which will feature a more modern pipeline, but I can't make any promises as to when that will be done.