@alert-judge-41842 werenyou able to resolve this ? i am having the same issue

Thanks — I can help get you unstuck. You said you’re seeing the same issue; are you asking for help reproducing the example, or are you stuck on a specific part of the flow? A few quick clarifying questions so I can give the exact code you need: • Which part is failing for you: the backend (Python agent/LLM/TTS), the LiveKit data publish step, or the React frontend rendering/playing audio? • Which Gemini TTS SDK or API are you using (Google Gemini HTTP API, a library, or something else)? If you paste the snippet you use to call GeminiTTS I can adapt the example to match. • Which OpenAI client code are you using in Python (openai package, OpenAI’s newer client, or a wrapper)? If you paste your current LLM call, I’ll show how to split formatted vs. spoken text. • How are you publishing data to the client now? Are you using the LiveKit Python server SDK’s

publish_data

(and which version), or sending via your app server (websocket/HTTP)? • Do you want the audio delivered as an uploaded file URL (S3/server) or streamed as an audio track via LiveKit? If you prefer, paste the minimal failing code (backend or frontend) and the exact symptom/error, and I’ll provide a patched example tailored to your stack (LiveKit publish call, GeminiTTS request, plus the React code to render/play).