Optimize: Check realtime mode early in _on_transcript

- Add early return if in realtime mode to skip LLM triggering logic
- In realtime mode, the LLM handles STT, turn detection, and responses itself
- Removes redundant check in else branch
- Improves code clarity and efficiency

Author: d3xvn

agents-core/vision_agents/core/agents/agents.py

@@ -767,6 +767,11 @@ async def _on_transcript(self, event: STTTranscriptEvent | RealtimeTranscriptEve
             self.conversation.complete_message(self._user_conversation_handle)
             self._user_conversation_handle = None
 
+        # In realtime mode, the LLM handles everything itself (STT, turn detection, responses)
+        # Skip our manual LLM triggering logic
+        if self.realtime_mode:
+            return
+        
         # Determine how to handle LLM triggering based on turn detection
         if self.turn_detection is not None:
             # With turn detection: accumulate transcripts and wait for TurnEndedEvent
@@ -784,7 +789,7 @@ async def _on_transcript(self, event: STTTranscriptEvent | RealtimeTranscriptEve
         else:
             # Without turn detection: trigger LLM immediately on transcript completion
             # This is the traditional STT -> LLM flow
-            if not self.realtime_mode and self.llm:
+            if self.llm:
                 self.logger.info(f"🤖 Triggering LLM response immediately (no turn detection)")
 
                 # Get participant from event metadata