diff --git a/CHANGELOG.md b/CHANGELOG.md
index 0f6af9e3f..2675da132 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -58,9 +58,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Deprecated
 
-- Function calls with parameters `(function_name, tool_call_id, args, llm,
-  context, result_callback)` are deprectated, use a single `FunctionCallParams`
-  parameter instead.
+- Function calls with parameters
+  `(function_name, tool_call_id, args, llm, context, result_callback)` are
+  deprectated, use a single `FunctionCallParams` parameter instead.
 
 - `TransportParams.camera_*` parameters are now deprecated, use
   `TransportParams.video_*` instead.
@@ -73,6 +73,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Fixed
 
+- Fixed an issue with HTTP Smart Turn handling, where the service returns a 500
+  error. Previously, this would cause an unhandled exception. Now, a 500 error
+  is treated as an incomplete response.
+
 - Fixed a TTS services issue that could cause assistant output not to be
   aggregated to the context when also using `TTSSpeakFrame`s.
 
diff --git a/src/pipecat/audio/turn/smart_turn/http_smart_turn.py b/src/pipecat/audio/turn/smart_turn/http_smart_turn.py
index 4f542f81d..7ca6c5673 100644
--- a/src/pipecat/audio/turn/smart_turn/http_smart_turn.py
+++ b/src/pipecat/audio/turn/smart_turn/http_smart_turn.py
@@ -40,7 +40,6 @@ class HttpSmartTurnAnalyzer(BaseSmartTurn):
     async def _send_raw_request(self, data_bytes: bytes) -> Dict[str, Any]:
         headers = {"Content-Type": "application/octet-stream"}
         headers.update(self._headers)
-        logger.trace(f"Sending {len(data_bytes)} bytes as raw body to {self._url}...")
         try:
             timeout = aiohttp.ClientTimeout(total=self._params.stop_secs)
 
@@ -66,7 +65,11 @@ class HttpSmartTurnAnalyzer(BaseSmartTurn):
                     error_text = await response.text()
                     logger.trace("Response Content (Error):")
                     logger.trace(error_text)
-                    response.raise_for_status()
+                    if response.status == 500:
+                        logger.warning(f"Smart turn service returned 500 error: {error_text}")
+                        raise Exception(f"Server returned HTTP 500: {error_text}")
+                    else:
+                        response.raise_for_status()
 
         except asyncio.TimeoutError:
             logger.error(f"Request timed out after {self._params.stop_secs} seconds")
@@ -76,5 +79,14 @@ class HttpSmartTurnAnalyzer(BaseSmartTurn):
             raise Exception("Failed to send raw request to Daily Smart Turn.")
 
     async def _predict_endpoint(self, audio_array: np.ndarray) -> Dict[str, Any]:
-        serialized_array = self._serialize_array(audio_array)
-        return await self._send_raw_request(serialized_array)
+        try:
+            serialized_array = self._serialize_array(audio_array)
+            return await self._send_raw_request(serialized_array)
+        except Exception as e:
+            logger.error(f"Smart turn prediction failed: {str(e)}")
+            # Return an incomplete prediction when a failure occurs
+            return {
+                "prediction": 0,
+                "probability": 0.0,
+                "metrics": {"inference_time": 0.0, "total_time": 0.0},
+            }