page

Phase 1 - Wake Word

Phase 1: Wake Word Detection

Goal

Get "Hey LARS" wake word working on a device.

Tasks

1. Train Custom Wake Word

  • [ ] Access openWakeWord Google Colab
  • [ ] Record/generate training samples for "Hey LARS"
  • [ ] Train model (~1 hour)
  • [ ] Export .tflite file

2. Test Detection

  • [ ] Install openWakeWord on local-ai server
  • [ ] Load custom model
  • [ ] Test with microphone input
  • [ ] Measure accuracy and false positives

3. Create Listener Service

  • [ ] Python script that runs continuously
  • [ ] Listens for wake word
  • [ ] On detection: trigger next phase
  • [ ] Systemd service for auto-start

Deployment Options

  1. local-ai server - Already has hardware
  2. Raspberry Pi - Dedicated always-on device
  3. Tablet/Phone app - PWA with microphone access
  4. Web browser - For testing

Success Criteria

  • "Hey LARS" detected reliably
  • < 1 second detection latency
  • Low false positive rate
  • Runs 24/7 without issues
ID: a2bd21f0
Path: LARS Voice Assistant > Implementation Roadmap > Phase 1 - Wake Word
Updated: 2025-12-30T19:41:33