A developer has spent 14 months building a fully local, open-source voice assistant that operates without any third-party cloud services. The technical stack integrates whisper.cpp for speech-to-text, llama.cpp or vLLM for the LLM, and Kokoro or Piper for text-to-speech. The project features a 5-second latency to the first spoken word and is released under AGPLv3 and Apache-2.0 licenses.