Guidance Needed: Setting Up 1:1 Call with Real-Time Transcription Using Call Automation SDK

[idrisfl]

Hi,

I am currently working on a project where I need to set up a 1:1 call between two individuals using the Azure Communication Services Call Automation SDK. My goal is to enable real-time transcription during the call. I have followed the basic setup instructions, but I am encountering some issues and would appreciate some guidance.

Here are the steps I have taken so far:

1. Created an Azure Communication Services resource.
2. Set up a WebSocket server to stream the transcription in real-time.
3. Established the call using the Call Automation SDK.
4. Configured the transcription options, including the language and WebSocket connection.

However, I am facing challenges with:

• Handling both IncomingCall and OutgoingCall events to capture details of both participants.
• Instead of the instance of an AI agent such as gpt-4o-realtime or whisper, the call receiver would be a human agent.
• I'm not able to obtain the individual participant Ids of both callers in the TranscriptionData object.

Could you please provide a detailed example or point me to a quickstart guide that covers these aspects? Any code snippets or additional documentation would be greatly appreciated.

Thank you for your assistance!