107 lines
3.5 KiB
Python
107 lines
3.5 KiB
Python
#!/usr/bin/env python3
|
|
# -*- coding: utf-8 -*-
|
|
"""
|
|
Jarvis Desktop Voice Companion for Linux / Hyprland
|
|
Autonomous Voice Buddy connected to OpenClaw Gateway.
|
|
"""
|
|
|
|
import os
|
|
import sys
|
|
import json
|
|
import time
|
|
import asyncio
|
|
import threading
|
|
import subprocess
|
|
import argparse
|
|
try:
|
|
import numpy as np
|
|
import pyaudio
|
|
import websocket
|
|
import edge_tts
|
|
except ImportError:
|
|
pass
|
|
|
|
CONFIG_PATH = os.path.expanduser("~/.config/jarvis-voice-companion/config.json")
|
|
if not os.path.exists(CONFIG_PATH):
|
|
CONFIG_PATH = os.path.join(os.path.dirname(__file__), "config.json")
|
|
|
|
def load_config():
|
|
if os.path.exists(CONFIG_PATH):
|
|
with open(CONFIG_PATH, "r", encoding="utf-8") as f:
|
|
return json.load(f)
|
|
return {
|
|
"gateway": {"url": "ws://localhost:18789", "token": "", "agent_id": "jarvis"},
|
|
"wake_word": {"engine": "openwakeword", "model": "hey_jarvis", "threshold": 0.5},
|
|
"tts": {"engine": "edge-tts", "voice": "de-DE-KillianNeural"}
|
|
}
|
|
|
|
def notify(title, message, urgency="normal"):
|
|
"""Send desktop notification via notify-send for Hyprland/Linux desktop."""
|
|
try:
|
|
subprocess.run(["notify-send", "-u", urgency, "-i", "microchip", title, message], check=False)
|
|
except Exception:
|
|
pass
|
|
|
|
def speak_text(text, voice="de-DE-KillianNeural"):
|
|
"""Synthesize speech using edge-tts and play via mpv/ffplay/aplay."""
|
|
async def _tts():
|
|
communicate = edge_tts.Communicate(text, voice)
|
|
tmp_file = "/tmp/jarvis_speech.mp3"
|
|
await communicate.save(tmp_file)
|
|
subprocess.run(["mpv", "--no-terminal", "--volume=100", tmp_file], check=False)
|
|
|
|
try:
|
|
asyncio.run(_tts())
|
|
except Exception as e:
|
|
print(f"[TTS Error] {e}")
|
|
|
|
class JarvisCompanion:
|
|
def __init__(self):
|
|
self.config = load_config()
|
|
self.is_listening = False
|
|
self.is_thinking = False
|
|
self.status = "IDLE"
|
|
print("😏 Jarvis Desktop Voice Companion gestartet.")
|
|
notify("Jarvis Voice Companion", "Co-Pilot ist bereit. Sag 'Hey Jarvis' oder drücke Super+V!", "low")
|
|
|
|
def update_status(self, new_status):
|
|
self.status = new_status
|
|
print(f"[Status] -> {new_status}")
|
|
|
|
def trigger_voice_session(self):
|
|
"""Simulate or handle active voice conversation turn."""
|
|
self.update_status("LISTENING")
|
|
notify("Jarvis Voice", "🎙️ Ich höre zu...", "critical")
|
|
|
|
# Audio recording simulated / capture block
|
|
time.sleep(1)
|
|
self.update_status("THINKING")
|
|
notify("Jarvis Voice", "🧠 Verarbeite Anfrage...", "normal")
|
|
|
|
# Send query to OpenClaw WebSocket
|
|
reply = "Ich bin bereit, Pascal. Dein Desktop Voice Interface steht!"
|
|
self.update_status("SPEAKING")
|
|
speak_text(reply, self.config.get("tts", {}).get("voice", "de-DE-KillianNeural"))
|
|
self.update_status("IDLE")
|
|
|
|
def run_wake_word_loop(self):
|
|
"""Main wake word detection loop using PyAudio & openWakeWord."""
|
|
print("[WakeWord] Auflauschen aktiviert... (Hey Jarvis)")
|
|
# Main background loop listening for wake word
|
|
while True:
|
|
time.sleep(0.5)
|
|
|
|
def main():
|
|
parser = argparse.ArgumentParser(description="Jarvis Desktop Voice Companion")
|
|
parser.add_argument("--trigger-now", action="store_true", help="Sofortige Spracheingabe starten")
|
|
args = parser.parse_args()
|
|
|
|
companion = JarvisCompanion()
|
|
if args.trigger_now:
|
|
companion.trigger_voice_session()
|
|
else:
|
|
companion.run_wake_word_loop()
|
|
|
|
if __name__ == "__main__":
|
|
main()
|