feat: Add STT service preload to prevent cold start delays

- Add _preload_stt_service() method to warm up STT model during VoiceCal startup
- Call /preload?model_size=base endpoint to trigger model loading
- Prevent WebSocket HTTP 503 errors caused by cold-starting STT service
- Add timeout handling and detailed logging for preload process
- Update to v0.8.3 with STT service preload warming

Resolves the timing issue where VoiceCal tries to connect before STT service is ready.

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>

core/mcp_audio_handler.py

@@ -131,6 +131,10 @@ class MCPAudioHandler:
             self.stt_http_url = self._find_working_http_endpoint(stt_http_urls, "STT")
             self.tts_http_url = self._find_working_http_endpoint(tts_urls, "TTS")
             
+            # Preload STT service to warm up the model and avoid cold start delays
+            if self.stt_http_url:
+                self._preload_stt_service()
+            
             # Check if we have any working STT service (WebSocket preferred, HTTP fallback)
             if self.stt_websocket_urls or self.stt_http_url:
                 print("🔧 WEBSOCKET/HTTP FALLBACK: STT service available")
@@ -143,6 +147,38 @@ class MCPAudioHandler:
         except Exception as e:
             print(f"🔧 WEBSOCKET/HTTP FALLBACK ERROR: {e}")
     
+    def _preload_stt_service(self):
+        """Preload STT service to warm up the model and avoid cold start delays."""
+        try:
+            import requests
+            import time
+            
+            preload_url = f"{self.stt_http_url}/preload?model_size=base"
+            print(f"🔧 STT PRELOAD: Warming up STT service at {preload_url}")
+            
+            start_time = time.time()
+            response = requests.get(preload_url, timeout=30)
+            elapsed_time = time.time() - start_time
+            
+            if response.status_code == 200:
+                result = response.json()
+                if result.get("status") == "success":
+                    print(f"✅ STT PRELOAD: Model loaded successfully in {elapsed_time:.1f}s - {result.get('message', 'Ready')}")
+                    logger.info(f"STT service preloaded successfully: {result.get('message')}")
+                else:
+                    print(f"⚠️ STT PRELOAD: {result.get('message', 'Unknown response')}")
+                    logger.warning(f"STT preload response: {result}")
+            else:
+                print(f"⚠️ STT PRELOAD: HTTP {response.status_code} - Service may still be starting")
+                logger.warning(f"STT preload failed: HTTP {response.status_code}")
+                
+        except requests.exceptions.Timeout:
+            print(f"⏰ STT PRELOAD: Timeout after 30s - Service may be cold starting, will try again later")
+            logger.warning("STT preload timed out - service may be cold starting")
+        except Exception as e:
+            print(f"⚠️ STT PRELOAD: Error warming up service - {e}")
+            logger.warning(f"STT preload error: {e}")
+    
     def _find_working_http_endpoint(self, urls: list, service_name: str) -> str:
         """Find working HTTP endpoint for fallback."""
         import requests

version.py

@@ -2,9 +2,9 @@
 Version information for ChatCal Voice-Enabled AI Assistant
 """
 
-__version__ = "0.8.2"
+__version__ = "0.8.3"
 __build_date__ = "2025-08-23T04:00:00"
-__description__ = "VoiceCal - Voice-Enabled AI Scheduling Assistant with WebSocket STT integration and Gradio fallback mechanisms"
+__description__ = "VoiceCal - Voice-Enabled AI Scheduling Assistant with STT service preload warming to prevent cold start delays"
 
 def get_version_info():
     """Get detailed version information"""