How to run inference on Whisper model endpoint using SDK?

[vaibhavsingh007]

Hi,
I am following this article to deploy a whisper model however, not sure how to run prediction using an audio file.

I tried searching the documentation but the only resources I found uses Sagemaker notebook with built-in functions to run inference using audio input.

How can I pass the audio file for inference using Python SDK?
Any help will be appreciated.