using openai's realtime model, how can I trigger a separate LLM (self-hosted) to generate speech and queue it before the realtime LLM? I'm using the realtime model for STT+LLM, but there's a race condition where my custom LLM is sometimes triggered before the realtime LLM, and sometimes after.

Thanks for reaching out, Anas. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.