Azure Custom Speech is a feature of Azure’s Speech service that allows you to enhance the accuracy of speech recognition for your specific applications and products. It allows models to be trained on your own data, which can be particularly useful if you have a specific domain or accent that is not well-represented in the general models provided by Azure. It can be used for real-time transcription, batch transcription, and translations.

I’ve been a software engineer now for 13 years now delivering many applications to production, used by millions of people. For 12 of those years, I’ve coded utilizing Test-Driven Development (TDD). It’s been instrumental in allowing me to deliver fast and with confidence. In this post, I’ll be covering high-level steps on what it is, why do it, and how to do it, with tips that I’ve learned through experience.

Hey there, I’m Adam!