Lip-Sync Video Kit
When to use: Building talking-avatar features: explainer avatars, multilingual dubbing, virtual presenters, and character lip-sync.
npx skills add lip-sync-video-kitgit clone https://github.com/skillhub-community/lip-sync-video-kit.gitDescription
A toolkit for building talking-head / lip-sync videos. The LipSyncPlayer takes an audio track + an avatar (a static image, a video, or a generated talking-head) and drives mouth animation synced to the audio — either via a cloud lip-sync API (adapter: fal, HeyGen, etc.) or an offline web-based approximation (viseme mapping driven by audio amplitude + phoneme hints). The preview player shows the result in a frame with the script overlay and subtitle toggle.
Workflow components: AvatarPicker (choose from supplied avatar images or upload a face), ScriptPanel (edit the narration text that generates the audio), VoicePicker (reuse TTS voices for the narration), and MultilingualDubbing (translate the script and generate synced dubs in other languages, with the lip-sync re-driven per language). Export: MP4 with burned subtitles, or WebM; captions exported as SRT. Includes a duration/length estimator and a 'natural pause' inserter so the avatar's timing feels human rather than robotic.
AI Hub
Need a model endpoint?
AI Hub is an OpenAI-compatible endpoint for Cursor, Claude Code, and Agent workflows.
Open AI Hub