Model · Talking presenters, Video

MuseTalk

A free, downloadable model that changes the lip movements of a face in a video to match new speech, fast enough to run in real time on a strong graphics card. MIT licence. Needs technical setup.

Go to MuseTalk’s website

At a glance

Open or closed
Fully open
Weights and code downloadable under a standard open licence
Code is MIT; the licence file doesn't say whether the weights are covered.
source · checked 12 Oct 2026 · checked, but the page was unclear or partial
Licence
MIT (Permissive (e.g. Apache 2.0, MIT))
Bundled components carry MIT, Apache 2.0 and BSD-3 licences; weights not mentioned.
source · checked 12 Oct 2026 · checked, but the page was unclear or partial
Can I sell what I make?
Yes
  • You must credit the maker
MIT has no commercial restriction; keep the copyright notice when you redistribute the software. Weight terms not stated.
source · checked 12 Oct 2026 · checked, but the page was unclear or partial
Cost
Free option: Free to download; you pay for your own hardware.
Plans below.
How you use it
On your own computer, Through another service
Gradio demo on Hugging Face Spaces.
source · checked 12 Oct 2026 · checked, but the page was unclear or partial
Skill needed
Technical
source · checked 12 Oct 2026 · checked, but the page was unclear or partial
Made by
Lyra Lab, Tencent Music Entertainment
No country stated on the pages read.
source · checked 12 Oct 2026 · checked on the maker's own page
Status
Available
Version 1.5 released 2025-03-28; latest dated news 2025-04-05.
source · checked 12 Oct 2026 · checked, but the page was unclear or partial

What you can make with it

audio to videovideo to video

Edits a 256x256 face region to match audio (Chinese, English, Japanese); 30fps+ on an NVIDIA V100. Takes a video or face image; talking_photo is an approximation.
source · checked 12 Oct 2026 · checked, but the page was unclear or partial

Plans and prices

No paid plans listed.

source · checked 12 Oct 2026 · checked on the maker's own page

Rights and safety

Real faces and voices
Copies a real face (any face video or photo), no consent step
No consent step or disclaimer about real faces on the README.
source · checked 12 Oct 2026 · checked, but the page was unclear or partial

Related

Where you can run it

Hugging Face

Other options for “A video, lips that match new audio”

Arcads, Dzine, EchoMimic, FantasyTalking, HunyuanVideo-Avatar, ImagineArt, InfiniteTalk, Kapwing, LatentSync, Mage Space, Magic Hour, Picsart, PixVerse, Rask AI, SadTalker, SeaArt, sync., TopView, VEED, Vozo, Wav2Lip

About this entry

Facts last checked 12 Oct 2026 against the maker’s own pages (each fact links to its source). Entry status: draft. Plain-English summary written by Plainly.