Audio2Blendshapes is a Blender add-on that generates facial lip-sync animations directly from audio clips placed in the Video Sequencer. It requires a mesh rigged with standard ARKit blendshapes and automatically retargets the resulting animation data to match the framerate of the current scene. All processing is performed locally on the user's machine without relying on external servers or subscriptions. The tool handles multilingual audio inputs, offering optimal performance for English and Mandarin Chinese speech.