diff --git a/src/pipecat/serializers/twilio.py b/src/pipecat/serializers/twilio.py
index 71ac60da5..5a5a9d0c2 100644
--- a/src/pipecat/serializers/twilio.py
+++ b/src/pipecat/serializers/twilio.py
@@ -13,7 +13,11 @@ from typing import Optional
 from loguru import logger
 from pydantic import BaseModel
 
-from pipecat.audio.utils import create_stream_resampler, pcm_to_ulaw, ulaw_to_pcm
+from pipecat.audio.utils import (
+    create_stream_resampler,
+    pcm_to_ulaw,
+    ulaw_to_pcm,
+)
 from pipecat.frames.frames import (
     AudioRawFrame,
     CancelFrame,
@@ -235,6 +239,10 @@ class TwilioFrameSerializer(FrameSerializer):
             deserialized_data = await ulaw_to_pcm(
                 payload, self._twilio_sample_rate, self._sample_rate, self._input_resampler
             )
+            if deserialized_data is None or len(deserialized_data) == 0:
+                # Ignoring in case we don't have audio
+                return None
+
             audio_frame = InputAudioRawFrame(
                 audio=deserialized_data, num_channels=1, sample_rate=self._sample_rate
             )
diff --git a/src/pipecat/services/stt_service.py b/src/pipecat/services/stt_service.py
index 5aebe10f3..104be4810 100644
--- a/src/pipecat/services/stt_service.py
+++ b/src/pipecat/services/stt_service.py
@@ -152,6 +152,13 @@ class STTService(AIService):
         else:
             self._user_id = ""
 
+        if not frame.audio:
+            # Ignoring in case we don't have audio to transcribe.
+            logger.warning(
+                f"Empty audio frame received for STT service: {self.name} {frame.num_frames}"
+            )
+            return
+
         await self.process_generator(self.run_stt(frame.audio))
 
     async def process_frame(self, frame: Frame, direction: FrameDirection):