AI-102 Microsoft Designing and Implementing a Microsoft Azure AI Solution Free Practice Exam Questions (2025 Updated)

The requirement is to extract structured information (invoice items, sales amounts, customer details) from scanned invoice images.

Form Recognizer (now part of Azure AI Document Intelligence) is specifically designed for extracting structured data from documents such as invoices, receipts, and business forms.

Custom Vision → used for image classification or object detection, not for text/data extraction.

Computer Vision → can extract text with OCR, but it doesn’t provide structured field-level extraction for invoices.

Immersive Reader → for accessibility (reading content aloud, translating text), not document data extraction.

Correct Answer: D

Box 1: Yes

The ActivityHandler.OnMembersAddedAsync method overrides this in a derived class to provide logic for when members other than the bot join the conversation, such as your bot's welcome logic.

Box 2: Yes

membersAdded is a list of all the members added to the conversation, as described by the conversation update activity.

Box 3: No

To allow users to cancel an in-progress transaction, you need an interruption mechanism.

In Bot Framework Composer, this is done using triggers such as an intent trigger or dialog trigger (e.g., a CancelIntent trigger that interrupts the current flow).

A language generator handles variations in bot responses.

A custom event is not required for simple cancellation.

A conversation activity handles messages or events but does not manage interruption logic.

Correct Answer: C

Code Recapstatic void GetKeyPhrases(TextAnalyticsClient textAnalyticsClient, string text)

{

var response = textAnalyticsClient.ExtractKeyPhrases(text);

Console.WriteLine("Key phrases:");

foreach (string keyphrase in response.Value)

{

Console.WriteLine($"\t{keyphrase}");

}

}

Then called with:

GetKeyPhrases(textAnalyticsClient, "the cat sat on the mat");

“The call will output key phrases from the input string to the console.”

Yes.

The method calls ExtractKeyPhrases, iterates over response.Value, and writes each key phrase to the console.

“The output will contain the following words: the, cat, sat, on, and mat.”

No.

The Text Analytics key phrase extraction returns meaningful phrases, not stop words or every word.

Likely result: "cat", "mat", maybe "sat" depending on language model. But not "the", "on".

“The output will contain the confidence level for key phrases.”

No.

The ExtractKeyPhrases method in Text Analytics API returns only key phrases as strings, not confidence scores.

Statement Analysis

The call will output key phrases from the input string to the console → Yes

The output will contain the following words: the, cat, sat, on, and mat → No

The output will contain the confidence level for key phrases → No

Final Answer

Azure Text Analytics – Key Phrase Extraction

TextAnalyticsClient.ExtractKeyPhrases

Microsoft References

When deploying Cognitive Services containers (such as Anomaly Detector), the container requires a billing parameter in the docker run command.

This parameter points to the Azure resource key and endpoint for billing.

Fluentd is for logging.

Http Proxy is optional for outbound connectivity.

Mounts are used for local storage but are not mandatory for billing.

Correct Answer: B

To restrict access so that only resources in vnet1 can access the Language service:

You configure the virtual network settings on the Language service resource.

This enables private endpoint integration with the VNet.

NSGs and Azure Firewall apply to subnets/resources but do not configure the service resource itself.

Containers can run Language service locally but are not needed here.

Correct Answer: C

In the Azure portal, the Keys and Endpoint blade for a Cognitive Service resource provides:

API keys (for authentication)

Endpoint URL (the REST interface base URL)

Identity blade relates to managed identities.

Properties blade shows metadata (resource ID, location, subscription ID).

Networking blade controls network access and firewall rules.

Correct Answer: B. Keys and Endpoint

You are developing a speech translation service for recording lectures given in English (United Kingdom). The goal is to provide transcripts in multiple target languages: English, French, Spanish, and German.

Key requirements from the code snippet:

Source Language (Recognition Language):

config.SpeechRecognitionLanguage = "en-GB";

The input is English (UK).

Target Languages:You need to define a list of supported target languages. Since English, French, Spanish, and German are supported, the correct syntax is:

var lang = new List() { "en", "fr", "de", "es" };

This uses ISO language codes expected by the Translator API ("fr" = French, "de" = German, "es" = Spanish).

Recognizer Type:Since this is speech translation, you must use:

using var recognizer = new TranslationRecognizer(config, audioConfig);

The TranslationRecognizer class is specifically for translating speech input into multiple languages.

Completed Codestatic async Task TranslateSpeechAsync()

{

var config = SpeechTranslationConfig.FromSubscription("69cad5cc-0ab3-4704-bdff-abf4aa07d85", "uksouth");

var lang = new List() { "en", "fr", "de", "es" };

config.SpeechRecognitionLanguage = "en-GB";

lang.ForEach(config.AddTargetLanguage);

using var audioConfig = AudioConfig.FromDefaultMicrophoneInput();

using var recognizer = new TranslationRecognizer(config, audioConfig);

var result = await recognizer.RecognizeOnceAsync();

if (result.Reason == ResultReason.TranslatedSpeech)

{

foreach (var element in result.Translations)

{

AppendToTranscriptFile(element.Value, element.Key);

}

}

}

First Blank: {"en", "fr", "de", "es"}

Second Blank: TranslationRecognizer

Final Answer (Answer Area Selections)

Speech Translation with Speech SDK

TranslationRecognizer class

Speech service supported languages

Microsoft References

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.

Explanation:

You are configuring speech-to-speech translation in an app. The workflow is:

Recognize spoken input language (in this case, English).

Translate the recognized text into the target language (in this case, German).

(Optional) Synthesize speech in the target language.

Step 1 – SpeechRecognitionLanguage

This property defines the input language for speech recognition.

Since the input is English, it should be set to "en-US".

Step 2 – AddTargetLanguage

This method specifies the output translation language(s).

Since the requirement is to translate into German, use "de".

Other options (incorrect here):

speechSynthesisLanguage → used if you want to output translated text as synthesized speech in the target language voice. Not required in this snippet.

voiceName → specifies the actual voice to use for speech synthesis, not translation.

Correct Completed Code Snippet:

var translationConfig = SpeechTranslationConfig.FromSubscription(SPEECH_SUBSCRIPTION_KEY, SPEECH_SERVICE_REGION);

translationConfig.SpeechRecognitionLanguage = "en-US";

translationConfig.AddTargetLanguage("de");

Speech Translation with Azure AI Speech SDK

SpeechTranslationConfig Class Reference

Microsoft References

The scenario involves a Language Understanding (LUIS) or Conversational Language Understanding (CLU) model that has:

An intent called FindContact (used for searching a contact list).

Training phrases like:

“Find contacts in London”

“Who do I know in Seattle?”

“Search for contacts in Ukraine”

Key RequirementThe model must recognize locations (London, Seattle, Ukraine) inside the FindContact intent.

The proposed solution is to create a new intent for location.

This is not correct because:

Intents represent what the user wants to do (the action, e.g., FindContact).

Entities represent details within the user’s utterance (the data, e.g., location).

Here, the intent is already defined: FindContact.

Location should be modeled as an entity (Location entity), not as a separate intent.

Why the Solution is IncorrectCreating a new intent for each location (London, Seattle, Ukraine) would fragment the model and make it harder to generalize. Instead, the model should capture location as a parameter of the single intent FindContact.

Keep the existing FindContact intent.

Add an entity for location (prebuilt GeographyV2 in LUIS or “Location” entity in CLU).

Train the model so that London, Seattle, Ukraine are recognized as entity values inside the FindContact intent.

Correct Approach

Correct Answer: B. No

Intents and entities in Language Understanding

Prebuilt entity: GeographyV2

Microsoft References

Scenario Recap:

You provision a QnA Maker service in a new resource group (RG1) with an App Service plan (AP1).

What gets created automatically?

When provisioning QnA Maker:

Azure App Service (to host the QnA Maker endpoint).

Azure Cognitive Search (to store and index the knowledge base).

Azure Storage (for logs, intermediate data).

Azure SQL Database (for storing KB content in structured form).

However, only two are mandatory/automatic in the scope of the exam context.

Azure App Service (E)

Azure Cognitive Search (D)

Correct Answer: D. Azure Cognitive Search and E. Azure App Service

Relational databases use SQL (Structured Query Language) to query and manipulate data.

B is incorrect: Relationships exist between tables (foreign keys, joins).

C is incorrect: Flexible schema is a characteristic of NoSQL, not relational.

D is incorrect: Relational databases aim to reduce duplicate data through normalization.

semi-structured data

The statement is:

"Varying fields for each entity in a JSON document is an example of ______."

Relational data → Stored in structured tables with fixed schema (e.g., SQL database). Not applicable here.

Structured data → Has a strict schema (columns and types are predefined). JSON allows variation, so not structured.

Semi-structured data → Data that does not require a fixed schema, but still uses markers (tags, keys) to separate elements. JSON and XML are classic examples. ✅

Unstructured data → Data without schema or labeling (e.g., videos, images, free text). Not JSON.

Since JSON documents can have varying fields per entity (flexible schema), this is considered semi-structured data.

Correct Answer: semi-structured data

Structured, semi-structured, and unstructured data

JSON in Azure

Microsoft References

The scenario involves storing text files, videos, audio streams, and virtual disk images, along with metadata and a unique ID. This exactly matches Object storage characteristics:

Stores large amounts of unstructured data (blobs).

Each object contains the data itself, metadata, and a unique identifier.

Azure Blob Storage is Microsoft’s object storage solution.

Other options:

Columnar → Used for analytics workloads, optimized for reading columns (e.g., data warehouses).

Key/Value → Good for simple lookups but not optimized for storing large multimedia files.

Document → Stores semi-structured JSON documents, not large binary objects.

Thus, the correct answer is Object storage.

Microsoft Reference: Azure Blob Storage (Object storage) overview

According to the course AI-102T00: Designing and Implementing a Microsoft Azure AI Solution1, the correct sequence of actions to run model1 on Host1 is:

Export model1 to Host1

Run the container

Configure disk logging

The explanation and references are as follows:

Export model1 to Host1: This step is required to deploy the custom speech-to-text model to the Docker host. You can use the Azure portal or the Azure CLI to export the model as a container image2.

Run the container: This step is required to start the container and run the model on the Docker host. You can use the Docker CLI to run the container image3.

Configure disk logging: This step is optional but recommended to monitor the performance and health of the container. You can use the Docker CLI to configure disk logging for the container4.

To make a chatbot use uploaded company data from the Chat playground (“use your data”), your code must:

Create a chat request (not a plain completion), so you need the ChatCompletionsOptions type for the request options.

This is the options class used for chat completions across Azure OpenAI SDKs and is the one that supports attaching data sources (extensions).

Configure the Azure AI Search data source as a chat extension so the model grounds answers on your indexed documents.

In the SDK, this is done by adding an AzureCognitiveSearchChatExtensionConfiguration entry under the extensions/dataSources collection and setting properties like SearchKey and IndexName. This is the specific configuration object for Azure AI Search integration with Chat Completions.

Therefore, the two dropdowns should be completed as:

First dropdown: ChatCompletionsOptions()

Second dropdown: AzureCognitiveSearchChatExtensionConfiguration

This matches how “Use your data (RAG)” is implemented in Azure OpenAI—chat request + Azure AI Search chat extension—so the bot can answer questions using your uploaded/company data. Microsoft Learn

Microsoft Azure AI Solution References

Azure OpenAI “Use your data” (concepts): how chat uses data sources like Azure AI Search. Microsoft Learn

Azure OpenAI client library for .NET (overview of chat options and extensions support). Microsoft Learn

Chat Completions Options (SDK docs, shows dataSources/extensions concept in chat options).

Examples/discussion of AzureCognitiveSearchChatExtensionConfiguration usage with SearchKey/IndexName.

=========================

In Azure Cognitive Search, a knowledge store is an optional capability that allows you to persist the enriched data from a skillset (enrichment pipeline) into an Azure Storage account. This is often used when you want to further analyze or visualize enriched content, e.g., sentiment analysis results from social media posts.

When defining a knowledge store, you need:

storageConnectionString

This is required so the enrichment pipeline can connect to your Azure Storage account.

Without it, the knowledge store cannot persist any data.

storageContainer

This specifies the container in Blob Storage where enriched documents or projections will be stored.

This allows you to persist the original documents and enriched outputs such as sentiment analysis scores.

Other options (not correct in this case):

tables → This is a projection type used within a knowledge store to structure output in Table Storage, but it’s not required in the definition itself.

files → Another projection type for storing enriched documents as files in Blob Storage, but again, not a required field in the definition itself.

objects → A projection type used for representing enriched documents as JSON objects in Blob Storage.

Thus, while tables, files, and objects are used inside projections within a knowledge store, the definition itself always requires a storage connection string and storage container.

Correct Answer:

A. storageContainer

C. storageConnectionString

Knowledge store in Azure Cognitive Search

How to define a knowledge store

Microsoft References