DRAG DROP
-
You are building a transcription service for technical podcasts.
Testing reveals that the service fails to transcribe technical terms accurately.
You need to improve the accuracy of the service.
Which five actions should you perform in sequence? To answer, move the appropriate actions from the list of actions to the answer area and arrange them in the correct order.
Show Answer
Correct Answer: Create a Custom Speech project.
Create a speech-to-text model.
Upload training datasets.
Train the model.
Deploy the model.
Explanation: To improve transcription of technical terms, you must use Azure Custom Speech: first create a Custom Speech project, then define a speech-to-text model, upload domain-specific training data, train the model on that data, and finally deploy the trained model for use.
Question 150
You have an Azure subscription that contains an Azure Cognitive Service for Language resource.
You need to identify the URL of the REST interface for the Language service.
Which blade should you use in the Azure portal?
A. Identity
B. Keys and Endpoint
C. Networking
D. Properties
Show Answer
Correct Answer: B
Explanation: In the Azure portal, the REST endpoint URL for an Azure Cognitive Service for Language is shown on the **Keys and Endpoint** blade. This blade displays both the service endpoint required for REST API calls and the access keys used for authentication.
Question 151
You are building an AI solution that will use Sentiment Analysis results from surveys to calculate bonuses for customer service staff.
You need to ensure that the solution meets the Microsoft responsible AI principles.
What should you do?
A. Add a human review and approval step before making decisions that affect the staff's financial situation.
B. Include the Sentiment Analysis results when surveys return a low confidence score.
C. Use all the surveys, including surveys by customers who requested that their account be deleted and their data be removed.
D. Publish the raw survey data to a central location and provide the staff with access to the location.
Show Answer
Correct Answer: A
Explanation: Decisions about employee compensation have significant personal and financial impact, so Microsoft’s responsible AI principles require human oversight, accountability, and fairness. Adding a human review and approval step ensures AI-driven sentiment analysis does not make automated, unchecked decisions that could be biased or incorrect. The other options either reduce data quality, violate privacy, or fail to address accountability.
Question 152
HOTSPOT
-
You are building a call handling system that will receive calls from French-speaking and German-speaking callers. The system must perform the following tasks:
• Capture inbound voice messages as text.
• Replay messages in English on demand.
Which Azure Cognitive Services services should you use? To answer, select the appropriate options in the answer area.
NOTE: Each correct selection is worth one point.
Show Answer
Correct Answer: To capture messages:
Speech-to-text
To replay messages:
Text-to-speech and Translator
Explanation: Inbound French and German calls must be transcribed to text, which requires Speech-to-Text. To replay messages in English, the text must be translated to English and then converted back to audio, which requires Translator and Text-to-Speech.
Question 153
SIMULATION -
Use the following login credentials as needed:
To enter your username, place your cursor in the Sign in box and click on the username below.
To enter your password, place your cursor in the Enter password box and click on the password below.
Azure Username:
-
Azure Password: XXXXXXXXXXXX -
The following information is for technical support purposes only:
Lab Instance: 12345678 -
Task -
You need to create and publish a Language Understanding (classic) model named 1u12345678. The model will contain an intent of Travel that has an utterance of
Boat.
To complete this task, sign in to the Language Understanding portal at httptc//www.luis-ai/.
Show Answer
Correct Answer: Sign in to https://www.luis.ai/
Create a new Language Understanding (classic) app named 1u12345678
Add an intent named Travel
Add an utterance: Boat
Train the model
Publish the model
Explanation: The task requires creating a LUIS (classic) application with a specific name, defining one intent (Travel) with a sample utterance (Boat), and then training and publishing the model so it becomes available for use.
Question 154
DRAG DROP
-
You have a Language Understanding solution that runs in a Docker container.
You download the Language Understanding container image from the Microsoft Container Registry (MCR).
You need to deploy the container image to a host computer.
Which three actions should you perform in sequence? To answer, move the appropriate actions from the list of actions to the answer area and arrange them in the correct order.
Show Answer
Correct Answer: From the Language Understanding portal, export the solution as a package file.
From the host computer, move the package file to the Docker input directory.
From the host computer, run the container and specify the input directory.
Explanation: Deploying a LUIS container requires exporting the trained app, placing the package where the container can read it, and then starting the container with that input. Building the container or retraining is not required after downloading the image from MCR.
Question 155
You are building a chatbot by using Microsoft Bot Framework Composer.
You need to configure the chatbot to present a list of available options. The solution must ensure that an image is provided for each option.
Which two features should you use? Each correct answer presents part of the solution.
NOTE: Each correct selection is worth one point.
A. an entity
B. an Azure function
C. an utterance
D. an adaptive card
E. a dialog
Show Answer
Correct Answer: D, E
Explanation: Adaptive cards are required to present a list of options with rich content such as images for each option. A dialog is used in Bot Framework Composer to structure the conversation flow and present that adaptive card at the appropriate point, as well as handle the user’s selection.
Question 156
Note: This question is part of a series of questions that present the same scenario. Each question in the series contains a unique solution that might meet the stated goals. Some question sets might have more than one correct solution, while others might not have a correct solution.
After you answer a question in this section, you will NOT be able to return to it. As a result, these questions will not appear in the review screen.
You are building a chatbot that will use question answering in Azure Cognitive Service for Language.
You have a PDF named Doc1.pdf that contains a product catalogue and a price list.
You upload Doc1.pdf and train the model.
During testing, users report that the chatbot responds correctly to the following question: What is the price of
A. Yes
B. No
Show Answer
Correct Answer: B
Explanation: In Azure Cognitive Service for Language question answering, issues with recognizing different ways of asking about prices are not solved by creating entities. The service relies on question–answer pairs and variations/synonyms. Improving coverage requires adding alternative phrasings (e.g., "price," "cost," "how much") or synonyms and retraining/publishing the knowledge base. Therefore, creating an entity for price is not the correct approach.
Question 158
DRAG DROP
-
You need to analyze video content to identify any mentions of specific company names.
Which three actions should you perform in sequence? To answer, move the appropriate actions from the list of actions to the answer area and arrange them in the correct order.
Show Answer
Correct Answer: Sign in to the Azure Video Analyzer for Media website.
From Content model customization, select Brands.
Add the specific company names to the include list.
Explanation: Company name detection in video is configured in Azure Video Analyzer under content model customization. You must access the service, choose the Brands model, and explicitly include the company names you want the analyzer to detect.
Question 159
You train a Conversational Language Understanding model to understand the natural language input of users.
You need to evaluate the accuracy of the model before deploying it.
What are two methods you can use? Each correct answer presents a complete solution.
NOTE: Each correct selection is worth one point.
A. From the language authoring REST endpoint, retrieve the model evaluation summary.
B. From Language Studio, enable Active Learning, and then validate the utterances logged for review.
C. From Language Studio, select Model performance.
D. From the Azure portal, enable log collection in Log Analytics, and then analyze the logs.
Show Answer
Correct Answer: A, C
Explanation: Before deployment, model accuracy can be evaluated using built-in evaluation outputs. The language authoring REST endpoint provides a model evaluation summary with metrics such as precision, recall, and F1 score. In Language Studio, the Model performance section shows these same evaluation metrics based on test data. Active Learning and log analysis rely on post-deployment user traffic, so they are not suitable for pre-deployment evaluation.
$19
Get all 278 questions with detailed answers and explanations
Instant download HTML + PDF delivered the moment payment clears.
Secure Stripe checkout we never see or store your card details.
7-day refund if files are defective see our refund policy.
