Supertone · Voice

Supertonic 3

Commercial use: conditions applyBigScience Open RAIL-M License

About 800 MB of VRAM, and the only model here licensed Open RAIL-M. RAIL permits commercial use but attaches a list of prohibited uses that you must pass through to everyone you redistribute to. It is also the only licence here under which non-consensual voice cloning is a licence violation; elsewhere it is at most a line in a model card.

Supertone/supertonic-3 on Hugging Face audio.cpp

The licence

Summarized from the linked source documents, last checked 25 August 2026. Every claim links to its source; anything unconfirmed is marked unconfirmed.

BigScience Open RAIL-M License

Commercial use: conditions apply

Open RAIL-M permits commercial use, but binds you and every downstream recipient to the use restrictions in its Attachment A, which you must pass through in any redistribution or derivative. The sample code is separately MIT.

Conditions

You must pass the Attachment A use restrictions through to anyone you redistribute the model or a derivative to.
Attachment A prohibits, among other things: violating applicable law; harming minors; generating false information intended to cause harm; generating content without disclaiming that it is machine-generated; defamation and harassment; impersonation or deepfakes without consent; automated decision-making affecting legal rights; discrimination; medical advice; and predictive-policing applications.

What about the audio you generate?

Explicit and in your favour: "Licensor claims no rights in the Output You generate using the Model. You are accountable for the Output you generate and its subsequent uses." The disclosure restriction still applies: generating content without disclaiming that it is machine-generated is a prohibited use.

Based on the licence text as of the date below. Licences change; the links carry the authoritative version.

Specs

Transcribed from the shipped catalog, audiocpp_models_catalog.json — that file is authoritative if a number here disagrees with it.

Vendor: Supertone
Catalog id: audiocpp-supertonic-3
audio.cpp loader family: supertonic
Can do: Text to speech
Weights repo: audio-cpp/audio.cpp-gguf
Weights file: Supertonic-3-GGUF/supertonic-3-q8_0.gguf
Account needed to download: No
Max clip length: 60s
Sample rate: 44.1kHz mono
Approx. VRAM: 0.8 GB
In the picker by default: No — opt in from Settings
Run end-to-end here: No — catalogued, not tested here

Getting it

There is no per-model install. One native binary runs all of these, so getting a model is a download of its weights and nothing else — no virtualenv, no torch pin, no CUDA wheel to match against your driver.

From the CLI

soundbox model install audiocpp
soundbox pull audiocpp-supertonic-3

The binary is fetched from audio.cpp's GitHub releases the first time you ask for it and cached under your own data directory. Soundbox ships no native binaries and vendors none.
Weights land in <models dir>/audiocpp/audiocpp-supertonic-3/ rather than in the shared Hugging Face cache, so a download stays scoped to the model you asked for.
This one is opt-in: switch it on in the Models page's “Enable more models” section before it appears in the picker.

Worth knowing

~800 MB of VRAM. Try this one first if your GPU is small.
Opt-in and untested here.

Sources

Primary sources: the model's own repository, the project it came from, and the conversion audio.cpp actually downloads. Never invented — an entry with nothing findable says so instead.

Licence details on these pages are summaries of the linked source documents, last checked 25 August 2026. They are not legal advice, and Soundbox is not affiliated with any model's authors — weights download from Hugging Face under the user's own account and terms. Company and model names identify whose work it is, and belong to their owners. The linked licence text is authoritative; commercial use should be checked against it, per model, before anything ships.