Model · Music and sound, Video

MMAudio

A research model that adds matching sound effects to a silent video clip, or makes sounds from a text description. Clips are about 8 seconds by default. Downloadable, but the model files are for non-commercial use only.

Go to MMAudio’s website

At a glance

Open or closed
Downloadable, with conditions
You can download and run it, but the licence limits use
Code is MIT; weights on Hugging Face are CC-BY-NC 4.0 (non-commercial).
source · checked 12 Oct 2026 · checked on the maker's own page
Licence
CC-BY-NC 4.0 (weights); MIT (code) (Non-commercial only)
source · checked 12 Oct 2026 · checked on the maker's own page
Can I sell what I make?
No
  • You must credit the maker
Weights are CC-BY-NC 4.0; the page also says the models may not be suitable for commercial use. Whether the licence covers generated sounds is not spelled out.
source · checked 12 Oct 2026 · checked, but the page was unclear or partial
Cost
Free option: Free to download and run yourself for non-commercial use
Plans below.
How you use it
On your own computer, Through another service
Only tested on Ubuntu. Demos linked on Hugging Face, Colab and Replicate.
source · checked 12 Oct 2026 · checked on the maker's own page
Skill needed
Technical
source · checked 12 Oct 2026 · checked on the maker's own page
Made by
University of Illinois Urbana-Champaign, Sony AI and Sony Group Corporation
Research collaboration; no single company location stated.
source · checked 12 Oct 2026 · checked on the maker's own page
Status
Available
source · checked 12 Oct 2026 · checked, but the page was unclear or partial

What you can make with it

video to audiotext to audio

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

Plans and prices

No paid plans listed.

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

Rights and safety

Limits
max audio seconds: 8; note: 8 seconds is the default and training length, not a hard cap; longer may lower quality.
source · checked 12 Oct 2026 · checked, but the page was unclear or partial

Related

Where you can run it

Hugging Face, Replicate

Other options for “A video, a soundtrack or sound for a video”

Adobe Premiere, Filmora, Magic Hour, PixVerse, Seedance, Sora, Sora 2, Veo 3, Wisecut

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.