To be honest, I’m still figuring it out.
Generally I’ll have a slider for the wiggle frequency and a slider for the amount, and I’ll always have a very slight, very quick wiggle going on (more of a stylistic choice inspired by the Funhaus YouTube channel’s style of editing), and then as the louder, more enunciated words come up, I’ll crank up the sliders for just those words, so the mouth opens up more. It doesn’t always work perfectly though, so I’m experimenting with ways to make it more streamlined.
I’m just now realizing that you can create key frames based on audio, so I wonder if I could link the mouth movement to the actual audio layer with some kind of linear function between the volume or pitch and the amount of mouth movement. I’ll have to look into that.
I’m also interested in that link you posted. I’ll check that out for sure. Thanks!