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

Merge pull request #2341 from getchannel/realtime-text

Browse Source 
Hotfix: Correct Gemini Live API class to fix 1007 payload error.
This commit is contained in:
Vanessa Pyne
2025-08-04 11:03:57 -05:00
committed by GitHub
parent 0184a8c231 258e83c904
commit 96b1000e52
1 changed files with 2 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
src/pipecat/services/gemini_multimodal_live/gemini.py
View File

@@ -983,17 +983,14 @@ class GeminiMultimodalLiveLLMService(LLMService):
with audio and video inputs, preventing temporal misalignment that can occur with audio and video inputs, preventing temporal misalignment that can occur
when different modalities are processed through separate API pathways. when different modalities are processed through separate API pathways.
After sending the text, we signal turn completion to trigger a model response For realtimeInput, turn completion is automatically inferred by the API based
for text-only interactions. on user activity, so no explicit turnComplete signal is needed.
Args: Args:
text: The text to send as user input. text: The text to send as user input.
""" """
evt = events.TextInputMessage.from_text(text) evt = events.TextInputMessage.from_text(text)
await self.send_client_event(evt) await self.send_client_event(evt)
# After sending text, we need to signal that the turn is complete.
evt = events.ClientContentMessage.model_validate({"clientContent": {"turnComplete": True}})
await self.send_client_event(evt)
async def _send_user_video(self, frame): async def _send_user_video(self, frame):
"""Send user video frame to Gemini Live API.""" """Send user video frame to Gemini Live API."""