- Deploy Music Assistant on Pi (10.0.0.199:8095) with host networking for
Chromecast mDNS discovery, Spotify + SMB library support
- Switch primary LLM from Ollama to Claude Sonnet 4 (Anthropic API),
local models remain as fallback
- Add model info tag under each assistant message in dashboard chat,
persisted in conversation JSON
- Rewrite homeai-agent/setup.sh: loads .env, injects API keys into plists,
symlinks plists to ~/Library/LaunchAgents/, smoke tests services
- Update install_service() in common.sh to use symlinks instead of copies
- Open UFW ports on Pi for Music Assistant (8095, 8097, 8927)
- Add ANTHROPIC_API_KEY to openclaw + bridge launchd plists
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Character schema v2: background, dialogue_style, appearance, skills, gaze_presets
with automatic v1→v2 migration. LLM-assisted character creation via Character MCP
server. Two-tier memory system (personal per-character + general shared) with
budget-based injection into LLM system prompt. Per-character TTS voice routing via
state file — Wyoming TTS server reads active config to route between Kokoro (local)
and ElevenLabs (cloud PCM 24kHz). Dashboard: memories page, conversation history,
character profile on cards, auto-TTS engine selection from character config.
Also includes VTube Studio expression bridge and ComfyUI API guide.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add full Pi 5 satellite setup with ReSpeaker 2-Mics pHAT for kitchen
voice control via Wyoming protocol. Includes satellite_wrapper.py that
monkey-patches WakeStreamingSatellite to fix three compounding bugs:
- TTS echo suppression: mutes wake word detection while speaker plays
- Server writer race fix: checks _writer before streaming, re-arms on None
- Streaming timeout: auto-recovers after 30s if pipeline hangs
- Error recovery: resets streaming state on server Error events
Also includes Pi 5 hardware workarounds (wm8960 overlay, stereo-only
audio wrappers, ALSA mixer calibration) and deploy.sh with fast
iteration commands (--push-wrapper, --test-logs).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Living room unit fully working: on-device wake word (hey_jarvis), voice pipeline
via HA (Wyoming STT → OpenClaw → Wyoming TTS), static PNG display states, OTA
updates. Includes deploy.sh for quick OTA with custom image support.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Combines homeai-character (service status, character profiles, editor) and
homeai-desktop (chat with voice I/O) into homeai-dashboard on port 5173.
- 4-page sidebar layout: Dashboard, Chat, Characters, Editor
- Merged Vite middleware: health checks, service restart, bridge proxy
- Bridge upgraded to ThreadingHTTPServer (fixes LAN request queuing)
- TTS strips emojis before synthesis
- Updated start.sh with new launchd service names
- Added preload-models to startup sequence
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Adds the desktop web assistant app (Vite + React) with OpenClaw bridge
proxy and exposes it on the local network (host: 0.0.0.0, port 5174).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Replace faster-whisper with wyoming-mlx-whisper (whisper-large-v3-turbo, MLX Metal GPU)
STT latency: 8.4s → 400ms for short voice commands
- Add Qwen3.5-35B-A3B (MoE, 3B active params, Q8_0) to Ollama — 26.7 tok/s vs 5.4 tok/s (70B)
- Add model preload launchd service to pin voice model in VRAM permanently
- Fix HA tool calling: set commands.native=true, symlink ha-ctl to PATH
- Add pipeline benchmark script (STT/LLM/TTS latency profiling)
- Add service restart buttons and STT endpoint to dashboard
- Bind Vite dev server to 0.0.0.0 for LAN access
Total estimated pipeline latency: ~27s → ~4s
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add HomeAI dashboard: service status monitor, character profile manager, character editor
- Add TTS voice preview in character editor (Kokoro via OpenClaw bridge → Wyoming)
- Custom preview text, loading/playing states, stop control, speed via playbackRate
- Fix Wyoming API breaking changes: remove `version` from TtsVoice/TtsProgram,
use SynthesizeVoice object instead of bare string in Synthesize calls
- Vite dev server proxies /api/tts and /api/health to avoid CORS issues
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Fix Wyoming satellite crash on wake word: convert macOS .aiff chimes to .wav
(Python wave module only reads RIFF format, not AIFF)
- Fix OpenClaw HTTP bridge: increase subprocess timeout 30s → 120s, add SO_REUSEADDR
- Fix HA conversation component: use HTTP agent (not CLI) since HA runs in Docker
on a different machine; update default host to Mac Mini IP, timeout to 120s
- Rewrite character manager as Vite+React app with schema validation
- Add Wyoming satellite wake word command, ElevenLabs TTS server, wakeword monitor
- Add Phase 5 development plan
- Update TODO.md: mark voice pipeline and agent tasks complete
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add openclaw-http-bridge.py: HTTP server translating POST requests to OpenClaw CLI calls
- Add launchd plist for HTTP bridge (port 8081, auto-start)
- Add install-to-docker-ha.sh: deploy custom component to Docker HA via SSH
- Add package-for-ha.sh: create distributable tarball of custom component
- Add test-services.sh: comprehensive voice pipeline service checker
Fixes from code review:
- Use OpenClawAgent (HTTP) in async_setup_entry instead of OpenClawCLIAgent
(CLI agent fails inside Docker HA where openclaw binary doesn't exist)
- Update all port references from 8080 to 8081 (HTTP bridge port)
- Remove overly permissive CORS headers from HTTP bridge
- Fix zombie process leak: kill child process on CLI timeout
- Remove unused subprocess import in conversation.py
- Add version field to Kokoro TTS Wyoming info
- Update TODO.md with voice pipeline progress
- Pull qwen2.5:7b model (~4.7GB) with native tool-calling support
- Configure OpenClaw to use qwen2.5:7b as primary model
- Fix HASS_TOKEN file (remove trailing comment)
- Verify tool calling works end-to-end with HA skill
- Test home-assistant skill: turn_on/turn_off lights
- Update TODO.md with completed Phase 4 tasks
- Add PHASE4_COMPLETION.md documentation
Tool calling now working:
✓ qwen2.5:7b returns proper tool_calls array
✓ OpenClaw parses and executes commands
✓ Home Assistant skill controls entities
✓ HA API connectivity verified
- home-assistant: controls lights, switches, media players, climate etc
via HA REST API at 10.0.0.199:8123; includes service/domain reference
- voice-assistant: voice-specific response style guide for TTS output
(concise, no markdown, natural speech)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Mac Mini M4 Pro is now the primary development machine.
Scripts handle both macOS (launchd, Metal) and Linux (systemd, CUDA/ROCm).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Root setup.sh orchestrator with per-phase dispatch (./setup.sh p1..p8 | all | status)
- Makefile convenience targets (make infra, make llm, make status, etc.)
- scripts/common.sh: shared bash library for OS detection, Docker helpers,
service management (launchd/systemd), package install, env management
- .env.example + .gitignore: shared config template and secret exclusions
P1 (homeai-infra): full implementation
- docker-compose.yml: Uptime Kuma, code-server, n8n
- Note: Home Assistant, Portainer, Gitea are pre-existing instances
- setup.sh: Docker install, homeai network, container health checks
P2 (homeai-llm): full implementation
- Ollama native install with CUDA/ROCm/Metal auto-detection
- launchd plist (macOS) + systemd service (Linux) for auto-start
- scripts/pull-models.sh: idempotent model puller from manifest
- scripts/benchmark.sh: tokens/sec measurement per model
- Open WebUI on port 3030 (avoids Gitea :3000 conflict)
P3-P8: working stubs with prerequisite checks and TODO sections
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Full project plan across 8 sub-projects (homeai-infra, homeai-llm,
homeai-voice, homeai-agent, homeai-character, homeai-esp32,
homeai-visual, homeai-images). Includes per-project PLAN.md files,
top-level PROJECT_PLAN.md, and master TODO.md.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
