Small Conversational speech models that can run on device

uv pip install -e .

Try on Gradio

python demo.py

• Vui.BASE is base checkpoint trained on 40k hours of audio conversations
• Vui.ABRAHAM is a single speaker model that can reply with context awareness.
• Vui.COHOST is checkpoint with two speakers that can talk to each other.

You can clone with the base model quite well but it's not perfect as hasn't seen that much audio / wasn't trained for long

vui is a llama based transformer that predicts audio tokens.

1. Was developed with on two 4090's https://x.com/harrycblum/status/1752698806184063153
2. Hallucinations: yes the model does hallucinate, but this is the best I could do with limited resources! :(
3. VAD does slow things down but needed to help remove areas of silence.

@software{vui_2025,
  author = {Coultas Blum, Harry},
  month = {01},
  title = {{vui}},
  url = {https://github.com/fluxions-ai/vui},
  version = {1.0.0},
  year = {2025}
}