KoljaB/mOrpheus_3B-1Base_early_preview-v1-25000_GGUF

mOrpheus_3B-1Base_early_preview (NSFW TTS)

A finetuned Orpheus text‑to‑speech model trained on adult data for more expressive sounds:
<laugh>, <chuckle>, <sigh>, <cough>, <sniffle>, <groan>, <yawn>, <gasp>
New in this model: <moans>, <panting>, <grunting>, <gagging sounds>, <chokeing>, <kissing noises>

Speaker name: baddy
Framework: Safetensors (LLaMA)
Status: Early preview; training still underway

---

🔗 Links

• Model files & versions: xet
• Discussion & bug reports: Discord server
• Original author: MrDragonFox

---

🚀 Usage (Example)

1. Load the *.GGUF file into LMStudio.

2. pip install RealtimeTTS[orpheus]
   

3. Play TTS:

   from RealtimeTTS import TextToAudioStream, OrpheusEngine
   
   engine = OrpheusEngine(model="morpheus_3b-1base")
   # or: engine = OrpheusEngine(model="orpheus_3b-1basegguf@q4_k_m")
   stream = TextToAudioStream(engine)
   engine.set_voice("baddy")
   
   stream.feed("Mmm <moans>... that feels so good <groan>")
   stream.play()
   

---

⚖️ License

This model is released under Creative Commons Attribution‑NonCommercial 4.0 International (CC‑BY‑NC‑4.0). That means:

• NonCommercial: You can use, convert, and share this model for non‑commercial purposes only.
• Attribution: You must credit MrDragonFox, include the license link, and note any changes you made.
• No extra restrictions: Don’t apply paywalls, DRM, or additional terms.

© 2025 MrDragonFox  
Licensed under [CC‑BY‑NC‑4.0](https://creativecommons.org/licenses/by-nc/4.0/)

---

⚠️ Disclaimer

• No warranties—use at your own risk.
• Still under development; results may vary.
• Please report bugs or suggestions on Discord.