PiperConvert text into speech locally using downloadable voice models.
Free and open source (GPL-3.0 licence per GitHub).
AI tools
LivePortrait
Free optionOpen sourceLocal
Animate a single portrait into a talking, expressive video locally with an open-weights model.
What you can use it for
Animate a single portrait into a talking, expressive video locally with an open-weights model.
- Covers the core everyday task of HeyGen and Synthesia: animating a portrait into a talking/lip-synced video from a still image or a driving video.
- Creators report using it for talking avatars and AI vtubing, and call its results among the best open-source lip-sync they have found.
Workflow coverage and comparison notes
Animate a still portrait into a talking, lip-synced video locally without a per-minute avatar subscription. Covers the core talking-head slice only; no stock-avatar library, script-to-video studio or hosted rendering, and the licence needs review for commercial use.
Before you switch
Not a managed studio: no browser editor, stock-avatar library or script-to-video pipeline, and setup takes manual work (ComfyUI nodes or CLI). Licensing needs care — the code has no standard SPDX id and the InsightFace dependency is non-commercial, though forks that swap it for MediaPipe (Apache-2.0) exist. Head rotation and long clips can deform; a capable GPU is recommended.
Alternatives to these paid apps
Each comparison covers the workflow described above, rather than every paid feature.
More for the same workflow
See alternatives →Piper
KokoroGenerate spoken audio from text using a local speech model.
Free and open source (Apache-2.0 licence per GitHub). Runs locally with no usage cap; you supply the computer.