cosyvoice2-0.5b (Vokra GGUF)

Text to speech model, 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
cosyvoice2-0.5b.gguf	2448.8 MB	bf4d5eb7d4be00118be4fa3c2605957e3699185d9dd1159a8710e6e8dd07c4c4

Usage

# Download (any HTTP client works — the file is a plain GGUF)
curl -L -o cosyvoice2-0.5b.gguf \
  https://huggingface.co/vokra/cosyvoice2-0.5b/resolve/main/cosyvoice2-0.5b.gguf

vokra-cli run --model cosyvoice2-0.5b.gguf --text "こんにちは" --output out.wav

Text input needs a G2P front-end; see the model notes below for which one this voice expects.

Provenance

Field	Value
Architecture	cosyvoice2
Tensors	295
Upstream source	FunAudioLLM/CosyVoice2-0.5B (Apache-2.0)
Upstream licence	Apache-2.0
Licence class	permissive
Registry model id	cosyvoice2
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.

Verifying this file

shasum -a 256 cosyvoice2-0.5b.gguf
# expect: bf4d5eb7d4be00118be4fa3c2605957e3699185d9dd1159a8710e6e8dd07c4c4