wx44wx/pipecat
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update comment in _maybe_mute_frame

Browse Source
This commit is contained in:
Mark Backman
2026-02-14 09:41:42 -05:00
parent c682a44bb6
commit 8f5e5e8e7c
1 changed files with 4 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
src/pipecat/processors/aggregators/llm_response_universal.py
View File

@@ -549,12 +549,10 @@ class LLMUserAggregator(LLMContextAggregator):
await s.cleanup() await s.cleanup()
async def _maybe_mute_frame(self, frame: Frame): async def _maybe_mute_frame(self, frame: Frame):
# Control frames must flow unconditionally — never feed them to mute # Lifecycle frames should never be muted and should not trigger mute
# strategies. Without this guard, strategies like # state changes. Evaluating mute strategies on StartFrame would
# MuteUntilFirstBotCompleteUserMuteStrategy fire on_user_mute_started # broadcast UserMuteStartedFrame before StartFrame reaches downstream
# and broadcast UserMuteStartedFrame before StartFrame is pushed # processors.
# downstream, causing downstream processors to receive frames before
# StartFrame and log errors.
if isinstance(frame, (StartFrame, EndFrame, CancelFrame)): if isinstance(frame, (StartFrame, EndFrame, CancelFrame)):
return False return False