smart-turn-v2 (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
smart-turn.gguf	361.6 MB	cb595fcc0197f6c9bc4110eea845ece1d3b966c13c2664bb056284434d7e494e

Usage

# Download (any HTTP client works — the file is a plain GGUF)
curl -L -o smart-turn.gguf \
  https://huggingface.co/vokra/smart-turn-v2/resolve/main/smart-turn.gguf

vokra-cli run --model smart-turn.gguf --input input.wav

Provenance

Field	Value
Architecture	smart_turn
Tensors	223
Upstream source	pipecat-ai/smart-turn-v2 (turn detection classifier, bsd-2-clause)
Upstream licence	bsd-2-clause
Licence class	permissive
Registry model id	smart-turn-v2
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 bsd-2-clause, unchanged from upstream. Conversion does not alter the licence, and your obligations run to the upstream author.

Verifying this file

shasum -a 256 smart-turn.gguf
# expect: cb595fcc0197f6c9bc4110eea845ece1d3b966c13c2664bb056284434d7e494e