SpeechBlend is a Unity package that provides real-time lip syncing for your characters. It works by analyzing audio from any Audio Source and using machine learning to predict realistic mouth shapes (visemes). The package supports various viseme blendshape sets, including Daz Studio, Character Creator 3, iClone, Adobe Fuse, and more. You can use SpeechBlend with a single jaw joint or mouth open blendshape for simple mouth tracking or with a character model that has viseme blendshapes for more realistic lip syncing. The package also supports live microphone input and offers various options to tweak the viseme prediction for optimal performance.
SpeechBlend works by analyzing audio from any Audio Source and using machine learning to predict realistic mouth shapes (visemes). It supports various viseme blendshape sets, including Daz Studio, Character Creator 3, iClone, Adobe Fuse, and more. The package can be used with a single jaw joint or mouth open blendshape for simple mouth tracking or with a character model that has viseme blendshapes for more realistic lip syncing. It also supports live microphone input and offers various options to tweak the viseme prediction for optimal performance.
