AlrecenkDid a bit more fine tuning on the avatar mouth mapping. In particular, I fixed an issue that added an extra syllable after you stop talking. I also improved the timing and recalibrated everything. #WorldFabric #VRM #avatar
It's still a bit jittery, but I think that's about as good as I can get on the audio processing. I could post-process the animation to make it appear more stable. I thought about dropping the smooth interpolation and going to a smaller number of fixed poses for a more anime style, but right now I'm focused on the engine features, not the specific style.
Nathan 🌻@Alrecenk I hope the question doesn't bother you, is the audio text to speech or not? I find it hard to tell ':)
If not please consider it a compliment to your diction ;)
@natpbs The audio is not text to speech, but I am using a voice changer. It's not my natural voice, but I don't think it changes my diction. I am deliberately speaking very clearly in this clip because it's a test of an audio processing pipeline. Need clear word to see how well the mouth matches.