Artificial intelligence (AI), is a set of tools that allow you to make predictive models that learn. Generally, AI application leverage three key factors:
- Ingesting data from multiple connected sources
- Building a model and training the data
- Deploying the model and tracking
The AI services in the cloud come in a few versions. Let’s take a quick look:
- Pre-built: These are as-a-service offerings. The types are as follows:
- Vision: Used to identify and analyze image content.
- Speech: To convert the spoken word to text.
- Language: To understand the meaning behind a speaker’s intent.
- Knowledge: Knowledge-based services.
- Search: Search-based services.
- Custom: It allows you to rapidly prototype your own model. Great information is available at https://docs.microsoft.com/en-us/azure/machine-learning/service/.
- Azure Machine Learning.
- Visual Studio Code Tools for AI.
- Machine Learning Studio.
- Conversational; the bot side of the world to naturally interact with folks.
- Bot Service: Interactive bot that can answer questions.
- Language Understanding
- QnA Maker: A question and answer maker
AI has a lot of real-world uses, such as predictive maintenance and defect detection. Bots also provide tons of interaction scenarios and you can use cognitive services with your apps, including SaaS solutions. I have used the speech service in D365 for dictation for a healthcare app and I have used search to rapidly search data repositories. It has also been used to enhance how developers write code.
So, when you want to build or leverage a cognitive service, let’s look at the dashboard:
You can see that there are a lot of services and ways to interact and leverage AI. Later in this book, we will discuss how to use these services when developing solutions.