Voicebox Model Collection

Voicebox is a collection of fine-tuned open-source language models designed for persona-driven dialogue. Each model is trained on custom datasets to adopt a distinct voice, vocabulary, and tone.


Available Models

Model Voice Status
Voicebox/cowboy Western / Cowboy Available
Voicebox/pirate Pirate Available
Voicebox/shakespeare Shakespearean Available
Voicebox/surfer Surfer Planned
Voicebox/knight Medieval Knight Planned
Voicebox/scientist Mad Scientist Planned
Voicebox/noir Noir Detective Planned
Voicebox/herald Town Herald Planned
Voicebox/butler Victorian Butler Planned
Voicebox/drill-sergeant Drill Sergeant Planned

Quick Start (Ollama)

  1. Create a Modelfile:
FROM ./cowboy.gguf

SYSTEM """
You are a cowboy assistant. Respond using western slang and character dialect.
"""
  1. Build the model:
ollama create cowboy -f Modelfile
  1. Run the model:
ollama run cowboy

Sample Responses

Prompt: "Tell me about the stars!"


Usage Notes


Maintained by QuillBytes. Feature requests and voice contributions can be submitted via repository issues.