cstr/gwen-tts-0.6b-GGUF

Gwen-TTS-0.6B — GGUF

GGUF conversion of Gwen-TTS-0.6B (a Qwen3-TTS variant) for the CrispASR qwen3-tts backend.

Multilingual text-to-speech with zero-shot voice cloning. Based on the Qwen3-TTS architecture: speech tokenizer + AR LM + flow-matching codec decoder.

Files

Usage

# Auto-download:
crispasr --backend qwen3-tts -m gwen-tts-0.6b-q4_k.gguf --auto-download \
    --tts "Hello, world." --tts-output out.wav

# With voice cloning:
crispasr --backend qwen3-tts -m gwen-tts-0.6b-q4_k.gguf \
    --voice reference.wav --ref-text "Reference transcript." \
    --i-have-rights \
    --tts "Text to synthesize." --tts-output cloned.wav

License

Apache 2.0.

Credits

• —GGUF conversion & runtime: CrispASR

Provenance and EU AI Act Art. 53 note

• —Upstream model: Gwen-TTS-0.6B (a Qwen3-TTS variant) — publisher not named on the card.
• —Upstream licence: apache-2.0. This repository redistributes under the same terms; it grants no rights the upstream licence does not.
• —What was done here: format conversion and/or quantisation only (GGUF). No training, no fine-tuning, no merging, no distillation, no change to architecture, vocabulary or capability. Only the numeric representation of the upstream weights differs.
• —Training data: documented — where it is documented at all — by the upstream provider; see the upstream model card. No training data was used, added or selected by this repository.
• —Provider status: under Regulation (EU) 2024/1689 the upstream authors remain the provider of this model. Converting the serialisation format does not make this repository the provider of a new general-purpose AI model, and no such claim is made. Questions about training content, copyright policy or model capability belong upstream.