CertPrep Coach

AI workloads

Recognize the common AI workloads, including text analysis, speech, computer vision, information extraction, and generative and agentic AI, and match each to a scenario.

Matching inputs to AI workloads Each kind of input maps to a workload: text to text analysis, audio to speech, images and video to computer vision, and documents and forms to information extraction. Generative AI and agents can work across all of them. Text reviews, emails, tickets Speech and audio calls, voice commands Images and video photos, camera feeds Documents and forms invoices, IDs, contracts Text analysis sentiment, entities, key phrases, summaries Speech speech to text, text to speech Computer vision describe, detect, read, generate images Information extraction fields and content from any file type Generative AI and agents: create content, reason and act across all inputs
Each kind of input maps to a workload: text to text analysis, audio to speech, images and video to computer vision, and documents and forms to information extraction. Generative AI and agents can work across all of them.

What Microsoft lists for this area

  • Identify scenarios for common AI workloads, including generative and agentic AI, text analysis, speech, computer vision, and information extraction
  • Describe common text analysis techniques, including keyword extraction, entity detection, sentiment analysis, and summarization
  • Identify features and capabilities of speech recognition and speech synthesis
  • Identify features and capabilities of computer vision and image-generation models
  • Identify techniques to extract information from text, images, audio, and videos

What this area tests

Given a business scenario, can you name the kind of AI it needs? Most questions describe what a company wants to do and offer workloads as answers. Start with the input (text, audio, images, documents), then the goal.

Text analysis

Text analysis pulls meaning out of written text:

  • Sentiment analysis: is the opinion positive, negative or neutral?
  • Entity detection: find named things such as people, places, organizations, dates and products.
  • Keyword (key phrase) extraction: list the main topics.
  • Summarization: shorten text into its key points.

Azure Language in Foundry Tools provides these as ready-made features. Microsoft now lists some of them, such as sentiment analysis and key phrase extraction, as legacy features kept for existing apps. Many of the same tasks can also be done by prompting a language model. For the exam, focus on what each technique does.

Speech

  • Speech recognition (speech to text): turn audio into a transcript.
  • Speech synthesis (text to speech): turn text into natural-sounding spoken audio.

Azure Speech in Foundry Tools provides both. A trap to watch: reading something aloud is synthesis, not recognition.

Computer vision and image generation

Computer vision analyzes images and video: describing a scene, detecting objects, reading printed text. Image generation works the other way, creating new images from a text prompt. Multimodal models can do both kinds of understanding as part of a conversation.

Information extraction

Information extraction pulls specific facts out of content: an invoice total, a name on an ID, the key moments in a recorded call. Azure Content Understanding handles documents, images, audio and video, and returns the results in a structure you define.

Generative and agentic AI

Generative AI creates new content: text, code, images or audio. Agentic AI goes a step further. An agent uses a model to reason about a request and then takes actions with tools, such as searching data or calling an API, to complete a task.

The difference matters on the exam. A chatbot that drafts replies is generative AI. A system that reads a support ticket, looks up the order and issues a refund is agentic.

Practice AI workloads

24 practice questions with explanations for this area.

Practice this area Or read 12 sample questions with answers

Sources

Updated October 25, 2026. Microsoft changes products often; check the official study guide for the latest outline.