audioseal-real-weight (Vokra GGUF)

Converted to the Vokra GGUF format for Vokra, a zero-dependency speech-AI inference runtime.

This is a conversion, not a new model. The weights are the upstream ones; Vokra re-packages them so its runtime can memory-map them directly. Credit for the model belongs upstream — see Source below.

Files

File	Size	SHA-256
audioseal-real-weight.gguf	178.0 MB	c44c5c4361ca38214a543931aea40baf7fa73939473f3eb568d67b1f6232f977

Usage

# Download (any HTTP client works — the file is a plain GGUF)
curl -L -o audioseal-real-weight.gguf \
  https://huggingface.co/vokra/audioseal-real-weight/resolve/main/audioseal-real-weight.gguf

vokra-cli run --model audioseal-real-weight.gguf --input input.wav

Provenance

Field	Value
Architecture	audioseal_real_weight
Tensors	310
Upstream source	facebook/audioseal (paired Generator + Detector 16-bit-message audio watermark for EU AI Act Article 50 compliance, San Roman et al. 2024 ICML arXiv:2401.17264, MIT — runtime binder gated on M5-05 T04 ADR ratification)
Upstream licence	mit
Licence class	permissive
Registry model id	audioseal_real_weight
Vokra GGUF schema	1
Converted by	vokra-core 0.1.0-alpha.0

Every row above is read out of this file's own vokra.* metadata, so the card cannot claim something the artifact does not carry.

Licence

The weights are distributed under mit, unchanged from upstream. Conversion does not alter the licence, and your obligations run to the upstream author.

Verifying this file

shasum -a 256 audioseal-real-weight.gguf
# expect: c44c5c4361ca38214a543931aea40baf7fa73939473f3eb568d67b1f6232f977