I am trying to create call from Python SDK to do polling so my agent must call people and starting talking first. By my workflow, agent must introduce something and start the poll But in reality it waits until I say something and only then starts saying first "say" node from the workflow. I have tried all combinations of "first_message_mode" and adding "first_message" but it does not work. Cam anyone help please? Really in a rush Here is the code I am trying to run await client.calls.create( customers=[ CreateCustomerDto( name=f"{request.poll_name}_{request.electorate}", number=p.number ) for p in request.phone_numbers ], assistant=CreateAssistantDto( name=f"{request.poll_name}_{request.electorate}", model=VapiModel( model="openai/gpt-4", workflow_id="2b2bdb78-ba28-406b-b640-6a46ace14edd", ), voice=VapiVoice( voice_id="Hana", provider="vapi", ), first_message_mode="assistant-speaks-first-with-model-generated-message", ), phone_number_id="61cabfff-0de8-4a9f-a82e-e80571b5c87a", )

Hey! To help track down this issue, could you share: - The call ID - When exactly this happened (the timestamp) - What response you expected to get - What response you actually got instead This would really help us figure out what went wrong!