Requesting Vertex AI Audio Output Allowlist for Project: pospinterpreter

Google Cloud Build with AI AI APIs
1

Hi Google Team,

We are an interpretation and translation services and technology company based in Finland (europe-north1). Through our platform (Youpret - [URL Removed by Staff], we intermediate thousands of human interpretations monthly, and we are now integrating AI-powered real-time translation and interpretation features directly into our application.

Our backend is deployed on Google App Engine (GAE) in the europe-west1 region, utilizing the unified google-genai Java SDK.

We want to utilize the new Gemini native audio-to-audio translation capabilities (requesting both TEXT and AUDIO modalities via responseModalities in generateContent). However, our requests are currently blocked with the following error:

“com.google.genai.errors.ClientException: 400 . You are not allowlisted to request audio output.”

Could you please add our Google Cloud project to the allowlist for native audio output generation?

Here are our project details:

  • GCP Project ID: (PII Removed by Staff)
  • Primary Region (Production): europe-west1 (EU - Belgium)
  • Secondary Regions: europe-west9 (EU - Paris) and us-central1 (US - Iowa)

Thank you so much for your support!

2

Hi ,
This looks like an access issue for native audio output, not an App Engine issue.

Since your use case is audio input → audio output, make sure you are using a Gemini Live API native-audio model and an endpoint that supports audio-to-audio, not just a regular generateContent flow with responseModalities: [TEXT, AUDIO].

The error:

You are not allowlisted to request audio output.

strongly suggests that the selected project is not enabled for audio output generation. If the model, endpoint, region, billing, APIs, and IAM permissions are all correct, this likely needs to be handled by Google Cloud Support or your Google Cloud account team. Community members usually cannot enable this from the forum.