Modelpublic

vokra/speechbrain-spkrec-ecapa-voxceleb

sourceHugging Faceapache-2.0updated 2mo agoView on Hugging Face

0likes39downloads

Model Card

speechbrain-spkrec-ecapa-voxceleb (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
`spkrec-ecapa-voxceleb.restamped.gguf`	79.4 MB	`f03292e93c215037b7d855281bb5786d600a1db3e51e9450c8caeb1360f1d571`

Usage

bash

# Download (any HTTP client works — the file is a plain GGUF)
curl -L -o spkrec-ecapa-voxceleb.restamped.gguf \
  https://huggingface.co/vokra/speechbrain-spkrec-ecapa-voxceleb/resolve/main/spkrec-ecapa-voxceleb.restamped.gguf

bash

vokra-cli run --model spkrec-ecapa-voxceleb.restamped.gguf --input input.wav

Provenance

Field	Value
Architecture	`ecapa_tdnn`
Tensors	200
Upstream source	speechbrain/spkrec-ecapa-voxceleb
Upstream licence	`apache-2.0`
Licence class	`permissive`
Registry model id	`speechbrain-spkrec-ecapa-voxceleb`
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 apache-2.0, unchanged from upstream. Conversion does not alter the licence, and your obligations run to the upstream author.

Attribution required

SpeechBrain (ECAPA-TDNN speaker embedding on VoxCeleb, apache-2.0)

This licence obliges you to display the attribution above when you ship something built on these weights. Vokra surfaces it at runtime via vokra_model_attribution (C ABI) and a CLI banner.

Verifying this file

bash

shasum -a 256 spkrec-ecapa-voxceleb.restamped.gguf
# expect: f03292e93c215037b7d855281bb5786d600a1db3e51e9450c8caeb1360f1d571