Jarvis — Offline Voice Coding Assistant
Say "Hey Jarvis", speak a request, and it is transcribed and handed to Aider wired to a local LLM as a one-shot coding command. Wake word, speech-to-text and the model all run on the machine, so nothing leaves it.
Overview
Say "Hey Jarvis", speak a request, and Jarvis transcribes it and hands it to Aider, which edits the chosen project using a local model. Everything runs on the machine; no audio or code leaves it.
The Problem
- Cloud assistants send your voice and your code to third parties.
- Typing every instruction to a coding agent breaks focus.
How It Works
- A single 16 kHz microphone stream is read in 80 ms frames.
- openWakeWord scores each frame; above the threshold, Jarvis starts listening.
- A few frames from just before the trigger are kept so the first word isn't clipped.
- Recording stops after 1.2 s of silence or 6 s at most.
- faster-whisper transcribes the audio locally.
- The text is sent to Aider as a one-shot command against the local llama-server model.
Highlights
- Fully offline and private.
- Same audio stream reused between wake word and command, so there's no gap.
- Start scripts for Windows and Linux/macOS.