AI-900: NLP Workloads

Topic snapshot

Sample questions

These questions are original IT Mastery practice items aligned to this topic area. They are designed for self-assessment and are not official exam questions.

Question 1

Topic: Describe Features of Natural Language Processing (NLP) Workloads on Azure

An online retailer wants to analyze customer feedback messages that are already stored as text. It needs to detect sentiment, identify city names mentioned in each message, and use a prebuilt Azure AI service instead of training a custom model. Which Azure service category should the company use?

Options:

• A. Azure AI Vision

• B. Azure AI Speech

• C. Azure OpenAI Service

• D. Azure AI Language

Best answer: D

Explanation: This is a text analytics requirement, not a speech or vision workload. Azure AI Language is the Azure service family designed for prebuilt NLP tasks such as sentiment analysis and entity recognition on written text.

The key is to match the workload to the service family. Here, the input is already written text, and the company wants built-in analysis for sentiment and named entities without training a custom model. Azure AI Language is the best fit because it provides prebuilt NLP capabilities for common text workloads.

Azure AI Speech is used when the main input or output is audio. Azure AI Vision is for images and video. Azure OpenAI Service is primarily for generative AI scenarios such as assistants, content generation, and prompt-based interactions, so it is not the most direct choice for standard text analytics needs.

For ready-made analysis of written text, Azure AI Language is the usual starting point.

• Speech mismatch Azure AI Speech fits spoken audio tasks such as speech-to-text, not direct analysis of typed feedback.
• Vision mismatch Azure AI Vision focuses on image and video workloads, so it does not meet the text analytics requirement.
• Generative overreach Azure OpenAI Service can process text, but it is not the simplest prebuilt choice for sentiment and entity extraction.

Question 2

Topic: Describe Features of Natural Language Processing (NLP) Workloads on Azure

A retail company is building a self-service kiosk in Azure. The kiosk must convert customers’ spoken requests into text and then read order pickup instructions aloud. Which Azure service should the company choose?

Options:

• A. Azure AI Language

• B. Azure AI Speech

• C. Azure OpenAI Service

• D. Azure AI Vision

Best answer: B

Explanation: This scenario requires two speech capabilities: turning spoken audio into text and turning text back into audio. Azure AI Speech is the Azure AI service designed for speech-to-text and text-to-speech workloads, so it best fits the kiosk requirement.

The core concept is matching the service to the workload type. Here, both the input and output are spoken audio, so the correct fit is Azure AI Speech, which supports speech recognition to transcribe spoken language and speech synthesis to generate audio from text.

In this scenario, the kiosk must:

• listen to a customer and create text
• take text instructions and speak them aloud

That maps directly to Azure AI Speech. Other Azure AI services may help with text analysis, image analysis, or content generation, but they do not directly satisfy the primary requirement to transcribe speech and produce spoken responses.

• Text analysis only Azure AI Language works with written text such as sentiment, entities, and key phrases, but the stem requires direct handling of audio.
• Vision mismatch Azure AI Vision is for images and video, not spoken requests or spoken responses.
• Wrong primary need Azure OpenAI Service is used for generative AI scenarios, but the decisive requirement here is speech recognition and speech synthesis.

Question 3

Topic: Describe Features of Natural Language Processing (NLP) Workloads on Azure

A company wants a solution that listens to a customer speaking in German and provides an English translation of the spoken message. Which Azure service category should the company use?

Options:

• A. Azure Machine Learning

• B. Azure AI Language

• C. Azure AI Speech

• D. Azure AI Vision

Best answer: C

Explanation: Azure AI Speech is the best fit because the input is spoken audio and the goal is translation between languages. It is designed for speech workloads such as speech recognition, speech synthesis, and speech translation.

The key concept is matching the service to the workload type. Translating speech between languages is a speech workload, so Azure AI Speech is the best Azure service category. It supports scenarios such as recognizing spoken words, converting text to speech, and translating spoken audio from one language to another.

Azure AI Language is mainly for understanding and analyzing text, such as sentiment analysis, entity extraction, and key phrase detection. Azure AI Vision is for images and video, not audio. Azure Machine Learning is used to build custom models, but for this fundamentals scenario you should choose the prebuilt Azure AI service that directly matches the requirement.

When the requirement centers on spoken input, Azure AI Speech is the clearest choice.

• Text-focused NLP The Azure AI Language option is tempting because it handles language workloads, but it is primarily for text analysis rather than spoken audio translation.
• Visual analysis The Azure AI Vision option applies to images and video, not speech.
• Custom model platform The Azure Machine Learning option is broader than needed when a prebuilt speech service already matches the scenario.

Question 4

Topic: Describe Features of Natural Language Processing (NLP) Workloads on Azure

A retail company is building a hands-free customer support kiosk in Azure. The solution must convert customers’ spoken questions into text, read answers back as natural-sounding audio, and use a prebuilt Azure AI service instead of custom model training.

Which service should the company choose?

Options:

• A. Azure AI Speech

• B. Azure AI Vision

• C. Azure OpenAI Service

• D. Azure AI Language

Best answer: A

Explanation: Azure AI Speech is the best fit because the kiosk must both transcribe spoken language and synthesize spoken responses. Those are core speech capabilities provided by a prebuilt Azure AI service.

This scenario is about a voice interface, not general text analysis, image analysis, or generative content creation. Azure AI Speech is designed for speech-to-text when users speak into the kiosk and text-to-speech when the system reads answers aloud. That makes it the most direct service match for a hands-free experience.

Azure AI Language works with text that already exists, such as analyzing sentiment or extracting key phrases. Azure AI Vision is for images and video. Azure OpenAI Service is for generative AI tasks such as drafting or summarizing content, but it is not the primary service for converting speech to text and text to spoken audio.

The key takeaway is to choose the service family that matches the input and output type: spoken audio points to Azure AI Speech.

• Text analysis only The option using Azure AI Language misses the requirement to handle spoken input and spoken output.
• Wrong modality The option using Azure AI Vision is for visual content such as images, not audio conversations.
• Generative mismatch The option using Azure OpenAI Service focuses on generating content, not core speech transcription and synthesis.

Question 5

Topic: Describe Features of Natural Language Processing (NLP) Workloads on Azure

A company will analyze supplier emails by using Azure AI Language. Which requirement is an example of entity recognition?

Options:

• A. Extract supplier names, city names, and shipment dates from emails.

• B. Translate each email from French to English.

• C. Produce a short summary of each email.

• D. Classify each email as positive, neutral, or negative.

Best answer: A

Explanation: Entity recognition is used when text must be scanned for named items and each item must be categorized, such as a person, organization, location, or date. Extracting supplier names, city names, and shipment dates from email matches that goal directly.

Entity recognition is an NLP task that finds specific items in text and labels what they are. Common entity categories include people, organizations, places, and dates. In this scenario, the goal is to locate supplier names, city names, and shipment dates inside email text, so entity recognition is the best match for the requirement. This is a standard capability of Azure AI Language for analyzing text content.

The key clue is the need to both find items in the text and classify each item by type.

• The sentiment-analysis option measures opinion or emotional tone, not named items.
• The translation option changes text from one language to another, not extracts entities.
• The summarization option condenses content, not identifies categorized names, places, or dates.

Question 6

Topic: Describe Features of Natural Language Processing (NLP) Workloads on Azure

A travel company is building a kiosk that must listen to customers speaking German and return the request in English in near real time. The team wants a prebuilt Azure AI service and does not want to train a custom model. Which Azure service category should they choose?

Options:

• A. Azure AI Speech

• B. Azure Machine Learning

• C. Azure AI Language

• D. Azure AI Vision

Best answer: A

Explanation: This scenario requires translating spoken audio from one language to another, which is a speech workload. Azure AI Speech is the prebuilt Azure service family designed for speech recognition, synthesis, and speech translation.

Azure AI Speech is the best fit when the input is audio and the system must recognize, synthesize, or translate speech. Here, the kiosk must listen to spoken German and provide the result in English, which matches speech translation at a fundamentals level. Because the company wants a prebuilt Azure AI service, it should use a service that already supports speech capabilities instead of building and training its own model.

Services centered on text analysis or image analysis do not match an audio-based translation requirement. The key clue is that the source content is spoken language, not text or images.

• Text analytics mismatch The language service is mainly for analyzing and understanding text, not translating live spoken audio.
• Vision mismatch The vision service works with images and video, such as OCR or object detection, not speech.
• Custom ML not needed The machine learning platform can build custom models, but it is not the best choice when a prebuilt speech service already fits the requirement.

Question 7

Topic: Describe Features of Natural Language Processing (NLP) Workloads on Azure

A company wants to create written transcripts from recorded customer calls in English. It does not need to translate the calls into another language or generate spoken audio. Which capability should it use?

Options:

• A. Speech translation

• B. Sentiment analysis

• C. Speech recognition

• D. Speech synthesis

Best answer: C

Explanation: Speech recognition is the capability that turns spoken words into text. In this scenario, the goal is only to transcribe English audio into English text, so translation and synthesized speech are not needed.

The core concept is the difference between converting speech to text, translating between languages, and generating audio from text. When a requirement says to take spoken audio and produce a written transcript in the same language, that is speech recognition.

Speech-related capabilities are commonly separated like this:

• Speech recognition: speech to text
• Speech translation: speech in one language to text or speech in another language
• Speech synthesis: text to spoken audio

Sentiment analysis is also different because it analyzes the opinion or emotion in text rather than performing transcription. The key clue here is “written transcripts” with no language change and no audio generation.

• Speech translation is tempting because it works with spoken input, but it changes the language rather than simply transcribing it.
• Speech synthesis does the reverse task by turning text into audio.
• Sentiment analysis examines meaning or tone in text, not raw speech-to-text conversion.

Question 8

Topic: Describe Features of Natural Language Processing (NLP) Workloads on Azure

A retail company stores thousands of customer review comments as text. The support team wants an AI solution that can automatically pull out the most important terms or short phrases from each comment, such as “battery life,” “late delivery,” and “return policy.” They do not need a sentiment score or a generated summary.

Which solution best fits this requirement?

Options:

• A. Use Azure OpenAI Service summarization.

• B. Use Azure AI Vision OCR.

• C. Use Azure AI Language key phrase extraction.

• D. Use Azure AI Language sentiment analysis.

Best answer: C

Explanation: The requirement is to find important terms and short phrases in existing text. In Azure AI Language, key phrase extraction is the NLP task built for that exact purpose, unlike sentiment analysis, summarization, or OCR.

Key phrase extraction is an NLP capability that identifies the main topics or important terms in text. In this scenario, the company already has customer comments in text form and wants phrases like “battery life” or “return policy” pulled from each review. That is a direct match for key phrase extraction in Azure AI Language.

This differs from nearby tasks:

• Sentiment analysis measures opinion, such as positive or negative tone.
• Summarization creates a shorter version of text.
• OCR reads printed or handwritten text from images.

The key takeaway is that when the goal is to surface the most important terms or phrases from text, key phrase extraction is the best fit.

• Sentiment mismatch measures whether the review is positive, negative, or neutral, not which phrases are most important.
• Summary mismatch generates condensed text, but the scenario asks for extracted phrases rather than a rewritten summary.
• Wrong service type OCR is for reading text from images or scanned documents, while the reviews are already stored as text.

Question 9

Topic: Describe Features of Natural Language Processing (NLP) Workloads on Azure

A company wants to use a prebuilt NLP capability to evaluate customer comments such as The setup took too long, but the final result was worth it. The goal is to label each comment as positive, neutral, or negative based on overall opinion, not a custom business category or a word list. Which NLP workload should the company use?

Options:

• A. Key phrase extraction

• B. Keyword matching

• C. Text classification

• D. Sentiment analysis

Best answer: D

Explanation: The requirement is to detect the overall opinion expressed in each comment. Sentiment analysis is the NLP workload that estimates whether text is positive, neutral, or negative by considering context rather than only exact words.

Sentiment analysis is a specific NLP workload for measuring the attitude or emotional tone in text. In customer feedback scenarios, it evaluates the full statement and predicts whether the opinion is positive, neutral, or negative, even when the wording is mixed or implied. In Azure, this is a prebuilt capability of Azure AI Language. That is different from general text classification, which assigns text to predefined categories such as billing, sales, or support. It is also different from simple keyword matching, which only checks whether certain words appear and can miss context or phrases like not bad. When the goal is opinion labeling, sentiment analysis is the most direct fit.

• Text classification groups text into chosen labels, but the stem asks for a built-in opinion-detection workload.
• Keyword matching depends on exact words and does not reliably capture context or mixed sentiment.
• Key phrase extraction surfaces important terms from text, not whether the tone is positive, neutral, or negative.

Question 10

Topic: Describe Features of Natural Language Processing (NLP) Workloads on Azure

A company stores thousands of recorded customer support calls. It needs an Azure service that can convert the spoken conversations into text so the calls can be searched and analyzed later. The company wants a prebuilt service and does not want to train a custom model. Which Azure service category best fits this need?

Options:

• A. Azure AI Vision

• B. Azure AI Speech

• C. Azure AI Language

• D. Azure Machine Learning

Best answer: B

Explanation: This is a speech recognition requirement because the input is recorded audio and the goal is text output. Azure AI Speech is the prebuilt Azure service designed for speech-to-text scenarios.

Choose Azure AI Speech when the source data is spoken audio and the business needs transcription or speech recognition. In this scenario, the company has recorded calls and wants written text so the content can be searched and used later for analysis. That is a direct speech-to-text workload. Because the requirement is for a ready-made Azure service rather than building and training a custom model, a prebuilt speech service is the best fit. Other Azure AI services may help after text exists, but they do not perform the initial conversion from audio to text. The key distinction is simple: audio input points to Azure AI Speech, while text input points to language analysis services.

• Text already exists the Azure AI Language option is for analyzing text such as sentiment or key phrases after transcription has already happened.
• Wrong input type the Azure AI Vision option works with images and OCR, not spoken conversations.
• Too much customization the Azure Machine Learning option is for building custom ML solutions, which is unnecessary when a prebuilt speech service fits the need.

Continue with full practice

Use the AI-900 Practice Test page for the full IT Mastery route, mixed-topic practice, timed mock exams, explanations, and web/mobile app access.

Try AI-900 on Web View AI-900 Practice Test

Related focused pages

• Free AI-900 Full-Length Practice Exam
• AI Workloads
• Machine Learning
• Computer Vision
• Generative AI

Free review resource

Read the AI-900 Cheat Sheet on Tech Exam Lexicon, then return to IT Mastery for timed practice.