MakeItTalk review
A flexible local tool for animating human and stylized portraits from speech.
Reviewed by iTechGuides Editors · Editorial team · Updated Oct 2026
MakeItTalk is a research implementation for generating talking-head videos from one portrait image and a speech audio signal. It is aimed at developers and researchers who want to run audio-driven facial animation locally on Linux. The system models speech content and speaker information separately, predicts speaker-aware facial landmarks, and produces expressive motion from a single image rather than requiring a full facial video dataset for each portrait.
Its standout strength is breadth across portrait styles. MakeItTalk supports natural human faces alongside paintings, sketches, cartoons, manga, and stylized caricatures, making it relevant for research involving varied visual identities rather than only conventional talking-head footage. Pretrained model components provide a starting point, while training scripts cover content and image-translation modules. A customized puppet workflow and Google Colab demos add practical paths for exploring the implementation, although the main workflow remains local and command-line based.
MakeItTalk is a focused research codebase, not a hosted commercial video service. Its Linux orientation, local animation scripts, and training components suit technical users who can manage a development workflow and want access to the underlying implementation. The Creative Commons Attribution-NonCommercial 4.0 license is appropriate for noncommercial research and experimentation, while commercial projects should choose a tool with licensing and service terms designed for that use. Pick MakeItTalk for portrait-style research and controllable local experimentation; choose an alternative if you need a hosted product, a commercial production workflow, or a broader platform ecosystem.
MakeItTalk pros and cons
- Where it wins
- Animates faces, cartoons, manga, sketches, paintings, and caricatures
- Separates speech content from speaker information for facial animation
- Includes pretrained components, training scripts, and Colab demos
- Where it doesn't
- Designed for local Linux workflows rather than hosted production use
- Command-line setup may require developer or research experience
- Creative Commons Attribution-NonCommercial 4.0 limits commercial use
MakeItTalk fact sheet, pricing and score →
Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.
Last updated · How we research and update