k2-fsa · Voice

OmniVoice

Non-commercial onlyCC-BY-NC (weights) / Apache-2.0 (code)

The project states that the code is Apache-2.0 and the pre-trained weights are CC-BY-NC because of constraints from their training data. audio.cpp's conversion table lists this package as Apache-2.0, which is the code's licence. You download the weights.

k2-fsa/OmniVoice 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.

CC-BY-NC (weights) / Apache-2.0 (code)

Non-commercial only

The pre-trained weights are CC-BY-NC (non-commercial only), which the project attributes to constraints from its training data. The code is Apache-2.0. Forks inherit the weights restriction.

Conditions

Non-commercial use of the weights only; attribution required.
The card prohibits "unauthorized voice cloning, voice impersonation, fraud, scams, or any other illegal or unethical activities".
The model card has no machine-readable licence field, so the CC-BY-NC statement is prose in the card body. It comes from the rights-holder, but it is not metadata.

What about the audio you generate?

No ownership or watermarking statement found on the card. No output terms could be confirmed either way.

Correction to the shipped catalog: The shipped catalog says "Unknown". The weights are non-commercial. audio.cpp's conversion table labels this package Apache-2.0. That is the code licence; following it would breach the weights' terms.

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: k2-fsa
Catalog id: audiocpp-omnivoice
audio.cpp loader family: omnivoice
Can do: Text to speech
Weights repo: audio-cpp/audio.cpp-gguf
Weights file: OmniVoice-GGUF/omnivoice-q8_0.gguf
Account needed to download: No
Max clip length: 60s
Sample rate: 44.1kHz mono
Approx. VRAM: 1.8 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-omnivoice

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-omnivoice/ 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

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.