Lip Sync is a Blender add-on from Charley3D that generates mouth animation directly from a voice recording. Load an audio clip, run the analysis, and the add-on extracts phonemes and writes keyframes for your character's mouth, so no manual keyframing is required. It can drive 3D shape keys or 2D spritesheet mouth shapes, and it integrates with Blender's Video Sequencer. Support for Pose Library animation is listed as upcoming.
Under the hood, the add-on combines Vosk for offline speech recognition, a custom build of eSpeak NG for converting text into timed phoneme sequences, and Phonemizer for phoneme segmentation across languages. More than 25 languages are supported, including English, French, Spanish, German, Russian, Hindi and Mandarin. On first use it downloads the Vosk model for the chosen language, then caches it locally for faster re-analysis and fully offline work afterward.
It runs on Windows, macOS (Intel and Apple Silicon) and Linux, and requires Blender 4.4 or newer. Lip Sync is released under the GNU General Public License v3.0 or later, and a video trailer accompanies the listing.