Jarvis-home is my fully local AI assistant running on NVIDIA Jetson Orin Nano. It combines speech-to-text, LLM reasoning, and text-to-speech with an always-improving on-device pipeline, and I actively push upgrades to it almost every day.

Core stack: Gemma 4 E2B (llama.cpp), whisper.cpp, Piper TTS, and local routing for intents like lights, weather, and vision. The project is designed to stay local-first with zero cloud dependency for assistant interaction.

Current Focus and Roadmap:

• Conversation memory across interaction paths
• Wake word flow (“Hey Jarvis”) on-device
• Better multimodal vision path for hardware mode
• Music and utility intents with faster routing