Share via

Azure Foundry GPT Real-time

Tutorvi Admin 0 Reputation points
2025-11-20T08:50:04.81+00:00

I’ve been struggling for several weeks to get the GPT real-time API working. The API versions referenced in the documentation often seem to mismatch, and I haven’t been able to find a clear, authoritative source that indicates which versions are correct. Because of this, I’m also having trouble getting the Azure AI Foundry real-time API to work.

Additionally, the following documentation link consistently returns 404 errors, even though it appears to be the official reference: https://learn.microsoft.com/en-us/azure/ai-services/speech-service/voice-live-quickstart?tabs=foundry-new%2Cmacos%2Ckeyless&pivots=programming-language-python

Azure AI Speech
Azure AI Speech
An Azure service that integrates speech processing into apps and services.
0 comments

1 answer

Sort by: Most helpful
Most helpful Newest Oldest
  1. Nikhil Jha (Accenture International Limited) 4,150 Reputation points Microsoft External Staff Moderator
    2025-11-21T09:53:33.7133333+00:00

    Hi Tutorvi Admin,

    First, it's important to clarify that there are two distinct real-time audio APIs in Azure, which may be causing some of your confusion:

    1. Azure OpenAI GPT Realtime API (WebRTC/WebSocket-based)​
    2. Azure AI Speech Voice Live API (Speech Service-based)​

    The documentation link you mentioned that returns a 404 error is attempting to reference the Voice Live API, not the GPT Realtime API.

    Here are the current official, Microsoft documentation links for the GPT Realtime API:

    1. GPT Realtime API for speech and audio - Quickstart - https://learn.microsoft.com/en-us/azure/ai-foundry/openai/realtime-audio-quickstart
    2. How to use the GPT Realtime API for speech and audio - https://learn.microsoft.com/en-us/azure/ai-foundry/openai/how-to/realtime-audio once you open this link you will find all the necessary documentation and answer to your confusion:User's image

    Note: Make sure to deploy in specified regions only.

    On the other hand, Azure AI Speech "Voice Live" API:

    • What it is: An integration that wraps the GPT model inside the Azure Speech service. It adds features like echo cancellation, custom neural voices, and advanced Voice Activity Detection (VAD).
    • Best for: Telephony, complex voice agents, or when you need specific Azure Speech features.
    • Correct Docs (New Link): Get started with Azure Speech Voice Live

    If this answers your query, kindly "Accept and upvote the answer" so it benefits the other community members.
    Happy to help! 😊

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.