The course is part of this learning path
This course provides a list of resources for some additional topics you should make sure you understand before taking the AI-100 exam, including:
- The speech-to-text API
- Term lists for Content Moderator
- Content Moderator Review tool
- LUIS app training, testing, publishing, and privacy
- Bot authentication
- Azure Machine Learning options
- IoT Edge
- Azure Cognitive Search
Congratulations on making it all the way through this learning path. If you’re preparing to write the AI-100 exam, there are some additional topics you should study. I’ll list them for you and show you where you can learn more about them.
Sometimes the speech-to-text API has a hard time understanding certain words or phrases that your organization uses a lot, which can be frustrating. You can help it figure out these words or combinations of words by giving it a list of phrases to look for in your audio. Go to this link for more details. To save you some typing, I put all of the links from this video in the transcript below.
Similarly, if you want Content Moderator to watch for particular terms that your organization finds objectionable, then you can add your list of terms by using the List Management API.
You should also understand how you can use the Review tool to allow people to check what the Content Moderator AI has flagged.
Once you’ve added entities, intents, and utterances to a LUIS app, you need to train it, test it, and publish it.
Since LUIS stores the conversations your users have with your bots, it’s important to know how to make your LUIS apps compliant with privacy regulations like GDPR.
If you have a bot that needs to access an external service, such as Office 365, then your bot will need to authenticate to that service. One common way to do that is to use a JSON Web Token. https://docs.microsoft.com/azure/bot-service/bot-builder-concept-authentication
If you have an artificial intelligence requirement that can’t be met by one of the Cognitive Services APIs, then you can build a custom AI service using Azure Machine Learning. For the exam, you don’t need to know how to create a machine learning model, but you do need to know what options Azure Machine Learning provides for compute targets and deployment targets. For example, if you need to deploy an AI service that can handle a high number of simultaneous requests very quickly, you should deploy it to an Azure Kubernetes Service cluster. Find out more at this link:
When you’re collecting data using digital devices such as cameras, it’s often more efficient to perform AI activities directly on the device than to send the data to the cloud for processing. For example, you might want to do object recognition directly on a security camera. Microsoft has a solution called IoT Edge that lets you deploy AI models to certain devices.
Azure Cognitive Search sounds like it must be part of Azure Cognitive Services, but it’s actually separate. It allows you to add a rich search capability to your applications.
That’s it for additional resources. If you have any questions or comments, please let us know.
Thanks and good luck on the exam!
About the Author
Guy launched his first training website in 1995 and he's been helping people learn IT technologies ever since. He has been a sysadmin, instructor, sales engineer, IT manager, and entrepreneur. In his most recent venture, he founded and led a cloud-based training infrastructure company that provided virtual labs for some of the largest software vendors in the world. Guy’s passion is making complex technology easy to understand. His activities outside of work have included riding an elephant and skydiving (although not at the same time).