OpenAI / Gemini 2.0 Streaming for PiDOg

Has anyone gotten or can point me how to get the pidog setup to use OpenAI streaming from either OpenAI or Gemini? I’m trying to get the performance of the camera / microphone to use multi-modal AI processing to speed up the time it takes for the dog to response to a situation.

From what i can tell in the python code the app is waiting for the entire response from the API before it then converts to speach.

We have not verified the requirement you mentioned.
We suggest you combine it with our source code and ask GPT if there are any good solutions to implement it.
Here is our source code: GitHub - sunfounder/pidog: Raspberry Pi Robot Dog from SunFounder