Utilising Azure Speech to Text Cognitive Services with PowerShell

Introduction

Yesterday I posted about using Azure Cognitive Services to convert text to speech. I also eluded that I’ve been leveraging Cognitive Services to do the conversion from Speech to Text. I detail that in this post.

Just as with the Text to Speech we will need an API key to use Cognitive Services. You can get one from Azure Cognitive Services here.

Source Audio File

I created an audio file in Audacity  for testing purposes. In my real application it is direct spoken text, but that’s a topic for another time.  I set the project rate to 16000hz for the conversion source file then exported the file as a .wav file.

The Script

The Script below needs to be updated for your input file (line 2) and your API Key (line 6). Run it line by line in VSCode or PowerShell ISE.

See the gist on github.

Summary

That’s it. Pretty simple once you have a reference script to work with. Enjoy.