Gemini 2.5 Flash Live API native audio -- can't get it to initiate the conversation

Gemini API
,
1

Has anyone figured out how to get Gemini Live Native Audio API to initiate a conversation?

The official Google guide here says “To have the Live API initiate the conversation, include a prompt asking it to greet the user or begin the conversation. Include information about the user to have the Live API personalize that greeting.”

But the following prompt doesn’t lead to the model initiating the conversation, any idea on how to make this work? :

**Persona:**
You are Laura, a career coach from Brooklyn, NY. You specialize in providing data driven advice to give your clients a fresh perspective on the career questions they’re navigating. Your special sauce is providing quantitative, data-driven insights to help clients think about their issues in a different way. You leverage statistics, research, and psychology as much as possible.
You only speak to your clients in English, no matter what language they speak to you in.

Conversational Rules:

  1. Introduce yourself: Warmly greet the client as soon as the conversation starts by saying “Hi, how’s it going?”. Do NOT wait for the user to initiate the conversation.

  2. Intake: After the greetings, ask for your client’s full name, date of birth, and state they’re calling in from.

  3. Discuss the client’s issue: Get a sense of what the client wants to
    cover in the session. DO NOT repeat what the client is saying back to them in your response. Don’t ask more than a few questions here.

For reference, I’m using gemini-2.5-flash-native-audio-preview-09-2025

2 Likes
2

I’m having the same problem. With the previous model, I was sending it an instruction as a first message. That doesn’t work with the new model either.

Also, I’m getting terrible results from the new gemini-2.5-flash-native-audio-preview-09-2025 in general: breaks up mid-sentence, gets stuck, speaks nonsense and so on. I’m worried that they’re considering this generally available.

4

Hi @Ak_Sh
Welcome to the AI Forum!!!

We apologize for the inconvenience. Could you please try again using version ≥1.43.0?
Let us know if you continue to experience any issues.

6

Hi Shivam, I am on 1.45.0

8

Hi @Ak_Sh

Could you please share the complete payload details along with the steps to reproduce the issue? This will help us investigate it more accurately and provide a more precise response.

10

Let’s look at this technical declaration to see whether it resonance to your request. The Google Generative Language API factually provides programmatical access to advanced AI models, allowing developers to integrate sophisticated text generation capabilities and build custom scientific synthesis platforms effectively.