From 53251dcb8883002a1679716c7cbdfa7bd834eccc Mon Sep 17 00:00:00 2001
From: Mark Backman <mark@daily.co>
Date: Thu, 23 Jan 2025 14:23:11 -0500
Subject: [PATCH] Add better error handling for OpenAIRealtimeBetaLLMService
 truncate errors

---
 CHANGELOG.md                                        |  4 ++++
 src/pipecat/services/openai_realtime_beta/openai.py | 13 ++++++++++++-
 2 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index a20a336c1..f63aeb2fb 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -14,6 +14,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Changed
 
+- Add defensive error handling for `OpenAIRealtimeBetaLLMService`'s audio
+  truncation. Audio truncation errors during interruptions now log a warning
+  and allow the session to continue instead of throwing an exception.
+
 - Modified `TranscriptProcessor` to use TTS text frames for more accurate assistant
   transcripts. Assistant messages are now aggregated based on bot speaking boundaries
   rather than LLM context, providing better handling of interruptions and partial
diff --git a/src/pipecat/services/openai_realtime_beta/openai.py b/src/pipecat/services/openai_realtime_beta/openai.py
index c1d7c4a4d..d5acf5afd 100644
--- a/src/pipecat/services/openai_realtime_beta/openai.py
+++ b/src/pipecat/services/openai_realtime_beta/openai.py
@@ -167,8 +167,11 @@ class OpenAIRealtimeBetaLLMService(LLMService):
         either the wall clock time or the actual audio duration to prevent invalid truncation
         requests.
         """
+        if not self._current_audio_response:
+            return
+
         # if the bot is still speaking, truncate the last message
-        if self._current_audio_response:
+        try:
             current = self._current_audio_response
             self._current_audio_response = None
 
@@ -179,6 +182,11 @@ class OpenAIRealtimeBetaLLMService(LLMService):
             elapsed_ms = int(time.time() * 1000 - current.start_time_ms)
             truncate_ms = min(elapsed_ms, audio_duration_ms)
 
+            logger.trace(
+                f"Truncating audio: duration={audio_duration_ms}ms, "
+                f"elapsed={elapsed_ms}ms, truncate={truncate_ms}ms"
+            )
+
             await self.send_client_event(
                 events.ConversationItemTruncateEvent(
                     item_id=current.item_id,
@@ -186,6 +194,9 @@ class OpenAIRealtimeBetaLLMService(LLMService):
                     audio_end_ms=truncate_ms,
                 )
             )
+        except Exception as e:
+            # Log warning and don't re-raise - allow session to continue
+            logger.warning(f"Audio truncation failed (non-fatal): {e}")
 
     #
     # frame processing