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

Update OpenAIRealtime image to video to align with GeminiLive

Browse Source
This commit is contained in:
Mark Backman
2026-01-08 13:17:07 -05:00
parent b90a34228f
commit 2aadac7a4d
4 changed files with 45 additions and 43 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
changelog/3360.added.md
View File

@@ -1,6 +1,8 @@
- Added image support to `OpenAIRealtimeLLMService` via `InputImageRawFrame`:
- New `start_image_paused` parameter to control initial image input state
- New `image_detail` parameter to set image processing quality ("auto", "low", or "high")
- `set_image_input_paused()` method to pause/resume image input at runtime
- `set_image_detail()` method to adjust image quality dynamically
- Automatic rate limiting (1 image per second) to prevent API overload
- New `start_video_paused` parameter to control initial video input state
- New `video_frame_detail` parameter to set image processing quality ("auto",
"low", or "high"). This corresponds to OpenAI Realtime's `image_detail`
parameter.
- `set_video_input_paused()` method to pause/resume video input at runtime
- `set_video_frame_detail()` method to adjust video frame quality dynamically
- Automatic rate limiting (1 frame per second) to prevent API overload