Commit Graph

4 Commits

Author SHA1 Message Date
Hitonabi 483eb0b2fb Feat: Hermes PC-Zugriff via MCP (executor + mcp_pc)
- mcp/mcp_pc.py: MCP-Server der PC-Tools an den Executor proxied
  (pc_shell, pc_screenshot, pc_type, pc_key, pc_open, pc_status)
- client/hermes-pc/executor.py: FastAPI auf Port 7777 (Daemon-Reg entfernt)
- client/hermes-pc/start.bat, install.bat, requirements.txt
- client/hermes-voice/agent_client.py, main.py: agent_client integration

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-26 23:32:18 +02:00
Hitonabi 8881d65c8d Fix: audio_output Hang + Stimme auf Seraphina
- speak() hing weil pygame.time.wait() die asyncio Event-Loop blockierte.
  Fix: time.sleep(0.05) statt pygame.time.wait() im Playback-Loop.
- asyncio.new_event_loop() statt asyncio.run() — thread-sicher, kein
  "event loop already running" in Worker-Threads.
- _speak_lock verhindert parallele TTS-Aufrufe.
- Stimme: de-DE-SeraphinaMultilingualNeural als Default (modern, weiblich,
  multilingual neural). Dropdown in Settings mit allen DE/EN Optionen.
- speak()-Signatur: voice + rate als Parameter (statt config-Import).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-26 22:25:53 +02:00
Hitonabi 037c4b11df Feat: Hermes Voice Client — CustomTkinter GUI + Push-to-Talk + .exe Build
- main.py: CustomTkinter GUI (400x520) mit Status-Indikator, Conversation-Log,
  Screenshot-Toggle und Settings-Dialog. Thread-sicherer UI-Queue für Hotkey-Callbacks.
- hotkey_listener.py: pynput globaler Hotkey (press/release) + KeyCapturer für
  interaktive Hotkey-Konfiguration im Settings-Dialog.
- config_manager.py: JSON-Settings in ~/.hermes-voice/settings.json mit DEFAULTS,
  load() merged gespeicherte mit Default-Werten.
- audio_input.py: Vereinfacht auf start_recording/stop_recording/transcribe (kein
  Wake-Word-Loop mehr, direkt hotkey-gesteuert).
- ai_client.py: Settings-dict statt config.py-Imports, MC2-URL/Modelle konfigurierbar.
- requirements.txt: customtkinter>=5.2.0 + pynput>=1.7.6 hinzugefügt, pystray entfernt.
- build.bat: PyInstaller --onefile --windowed → dist/HermesVoice.exe.
- setup.bat: Veraltete Wake-Word-Hinweise entfernt, GUI-Check angepasst.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-26 22:23:36 +02:00
Hitonabi e19831f7d5 Feat: Hermes Voice Client (Wake Word + STT + Vision + TTS)
Windows-Desktop-Script: Energy VAD + Whisper Wake Detection,
faster-whisper STT, mss Screen Capture, MC2 Vision/Chat API,
Edge TTS Ausgabe, pystray System Tray. Kein Account nötig —
Wake Word frei konfigurierbar via WAKE_WORDS in config.py.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-26 22:02:52 +02:00