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

Fix RTVIObserver missing transcriptions from Realtime LLMs

Browse Source 
Realtime LLMs (OpenAI Realtime, Gemini Live) push TranscriptionFrame
and InterimTranscriptionFrame upstream only, but the RTVIObserver
filtered to downstream-only frames. Allow transcription frames
regardless of direction so user transcriptions are reported to
RTVI clients in all pipeline configurations.
This commit is contained in:
Mark Backman
2026-02-14 17:26:38 -05:00
parent c682a44bb6
commit 8d5d8219da
1 changed files with 9 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
src/pipecat/processors/frameworks/rtvi.py
View File

@@ -1220,11 +1220,15 @@ class RTVIObserver(BaseObserver):
frame = data.frame frame = data.frame
direction = data.direction direction = data.direction
# Only process downstream frames. Some frames are broadcast in both # Many frames are broadcast in both directions (e.g.
# directions (e.g. UserStartedSpeakingFrame, FunctionCallResultFrame), # UserStartedSpeakingFrame, FunctionCallResultFrame), so we default
# and we only want to send one RTVI message per event. # to only processing downstream copies to avoid duplicate RTVI
if direction != FrameDirection.DOWNSTREAM: # messages. However, some frame types (like transcriptions) may only
return # flow upstream (e.g. from Realtime LLMs), so we allow those
# regardless of direction.
if not isinstance(frame, (TranscriptionFrame, InterimTranscriptionFrame)):
if direction != FrameDirection.DOWNSTREAM:
return
# If we have already seen this frame, let's skip it. # If we have already seen this frame, let's skip it.
if frame.id in self._frames_seen: if frame.id in self._frames_seen: