Back to Projects
Project v0.4.2

HomeLlama: Voice Assistant

A fully local voice assistant running on a single NVIDIA Jetson Orin.

Tech Stack

Whisper.cpp for STT Llama-3 8B Quantized HomeAssistant API integration
View Repository Live Demo

About

HomeLlama is a privacy-focused voice assistant that runs entirely on local hardware. No cloud services required!

Features

  • Voice-to-text using Whisper.cpp
  • Natural language processing with Llama-3
  • Smart home integration via HomeAssistant
  • Fully offline operation