Is it possible to send real-time data to Azure Speaker Recognition API?

I am writing a voice-related project. I am developing a function to distinguish the voices of the people in the voices in real time. I use the Microsoft Azure Speaker Recognition API to distinguish people's voices.

According to the API guide, I have to upload a WAV file to receive audio and Differentiate the users in the voice. However, I need to using real-time audio streams to distinguish users voice.

So my question is : how do I use a real-time audio stream to implement Speaker Recognition?

  1. Can I get real-time audio using Azure Speaker Recognition API, which requires only WAV files?
  2. Or is there any other way than this API?