Free Practice Questions for Microsoft AI-102 Exam

You are using the Azure OpenAI .NET SDK to send a prompt (“What is Microsoft Azure?”) to a text model deployment. For text generation, you must call the Completions endpoint; hence client.GetCompletions(...) (or the async variant) is correct.

The Response < Completions > object returns one or more choices; the generated text is contained in response.Value.Choices[0].Text , which you can write to the console.

GetEmbeddings would return vector embeddings, not natural-language text, and GetImageGenerations is for DALL·E image generation, so neither meets the requirement to “answer user questions” and print text.

Microsoft References

Azure OpenAI Service (.NET) – Completions quickstart and sample showing GetCompletions and reading Choices[0].Text .

Azure OpenAI Service – Embeddings overview (returns vectors, not text).

Azure OpenAI Service – Image generation overview (DALL·E, not for text answers).

For the category field in Azure Cognitive Search:

searchable → allows users to search by words in the category field.

facetable → enables drill-down filtering (facets).

retrievable → ensures the field appears in search results.

filterable/sortable/key are not required here based on the scenario.

Correct Answer: A

The requirement:

You are using Azure AI Anomaly Detector.

You need to deploy it on Server1, which has intermittent internet access.

This means you need to run the service locally using containers (Docker).

Step 1 – Install Docker Engine

Before pulling or running any containers, Docker must be installed on Server1.

Without Docker Engine, no container can be deployed.

Step 2 – Pull the container image

Use docker pull to download the Anomaly Detector container image from Microsoft Container Registry.

This requires internet access, so it should be done when connectivity is available.

Step 3 – Run the container

After pulling the image, use docker run to start the Anomaly Detector container.

This exposes a local REST API endpoint on Server1.

Step 4 – Query the local endpoint

Once the container is running, you can query the prediction endpoint on Server1.

This allows local inference without requiring continuous internet access.

Correct Sequence:

Install the Docker Engine on Server1.

From Server1, run the docker pull command.

From Server1, run the docker run command.

Query the prediction endpoint on Server1.

Microsoft References

Use Azure AI Anomaly Detector on containers

Deploy Azure AI services in Docker containers

Anomaly Detector container setup

 Service endpoint → Computer Vision analyze images

 Property → imageType

The task:

" You have a library with thousands of images. You need to tag the images as photographs, drawings, or clipart. "

Which service endpoint?

Computer Vision analyze images → Provides a broad analysis including categories, tags, descriptions, and image type (photo, drawing, clipart). ✅

Computer Vision object detection → Identifies objects (people, cars, animals), not image type.

Custom Vision image classification / object detection → Requires custom training, not needed since the built-in Computer Vision service already supports this.

Correct: Computer Vision analyze images

Which property?

categories → Provides predefined categories (e.g., outdoor, indoor, people), not specifically drawing/clipart/photo.

description → Generates captions in natural language, not classification into photo/drawing/clipart.

imageType → Identifies if an image is a photograph, line drawing, or clipart. ✅

metadata → Provides size, format, dimensions.

objects → Lists detected objects, not image type.

Correct: imageType

Correct Answer:

Service endpoint → Computer Vision analyze images

Property → imageType

Microsoft References

Computer Vision Analyze API

ImageType property in Computer Vision

Creating an entity for price is a feature of Conversational Language Understanding (CLU/LUIS), not Question Answering. Custom question answering does not use entities for extraction or intent resolution; it matches user utterances to Q/A pairs using alternates, synonyms, and ranking. Therefore, defining an “entity for price” would not make the chatbot correctly answer “How much does < product > cost?”. The correct fix is to author additional alternate phrasings (as in Q266).

References

Custom question answering capabilities (no entity modeling).

https://learn.microsoft.com/azure/ai-services/language-service/question-answering/overview

CLU entities vs. QnA authoring—separate features and workflows.

https://learn.microsoft.com/azure/ai-services/language-service/conversational-language-understanding/overview

For a phone-style bot that listens to customers and replies with spoken output, you use two core Azure AI Speech SDK objects:

SpeechRecognizer performs speech-to-text (STT), turning the caller’s audio into text that your app can use to query the orders database. This minimizes effort because it handles audio capture/streaming, language models, and returns recognized text events/results directly.

SpeechSynthesizer performs text-to-speech (TTS), converting the retrieved order status text into natural-sounding audio to play back to the customer. It supports neural voices and simple calls like SpeakTextAsync() .

Other options are not appropriate here:

TranslationRecognizer is for real-time speech translation, not needed for simple STT.

VoiceProfileClient is for speaker verification/identification, not for recognition or synthesis of the conversation content.

Microsoft References

Azure AI Speech SDK – Speech to text (SpeechRecognizer) overview and quickstarts.

Azure AI Speech SDK – Text to speech (SpeechSynthesizer) overview and quickstarts.

Azure AI Speech SDK concepts – recognition vs. synthesis pipelines.

By default, Azure AI Language may log request and response data for monitoring.

Setting the loggingOptOut parameter to true ensures that input data is not persisted after analysis.

piiCategories specifies which PII types to detect.

showStats enables request/response statistics.

model-version specifies which model version to use.

Correct Answer: A

You can use the Speech service to voice-enable a chat bot.

The Direct Line Speech channel uses the text-to-speech service, which has neural and standard voices.

You ' ll need to make a small configuration change so that your bot can communicate with the Direct Line Speech channel using web sockets.