7.6. Practice Questions: NLP

Test your understanding of natural language processing and Azure AI Language/Speech services.

Domain 4: Natural Language Processing (Questions 31-40)

Question 31: Which NLP capability identifies the emotional tone of text?

• A. Key phrase extraction
• B. Entity recognition
• C. Sentiment analysis ✓
• D. Language detection

Rationale: Sentiment analysis determines whether text is positive, negative, or neutral.

Question 32: Which service provides text-to-text translation?

• A. Azure AI Speech
• B. Azure Translator ✓
• C. Azure AI Language
• D. Azure AI Vision

Rationale: Azure Translator supports text-to-text translation in 60+ languages. It does NOT support speech.

Question 33: What confidence score does language detection return for an unknown language?

• A. 0
• B. 1
• C. -1
• D. NaN ✓

Rationale: When language is unknown, the confidence score is NaN (Not a Number), and the language name is "Unknown."

Question 34: Which THREE values does language detection return?

• A. Bounding box coordinates
• B. Language Name ✓
• C. ISO 639-1 Code ✓
• D. Confidence Score ✓

Rationale: Language detection returns the language name, ISO code, and confidence score—not bounding boxes (that's vision).

Question 35: Which feature provides Wikipedia links for extracted entities?

• A. Key phrase extraction
• B. Entity linking ✓
• C. Named entity recognition
• D. Sentiment analysis

Rationale: Entity linking connects recognized entities to Wikipedia entries for additional context.

Question 36: Which Azure AI Speech feature can identify distinct speaker voices?

• A. Language identification
• B. Speech synthesis
• C. Speech recognition ✓
• D. Text-to-speech

Rationale: Speech recognition includes speaker recognition, which maps voice patterns to distinct speakers.

Question 37: Which THREE features are part of Azure AI Speech service?

• A. Document translation
• B. Language identification ✓
• C. Speaker recognition ✓
• D. Voice assistants ✓

Rationale: Language identification, speaker recognition, and voice assistants are Speech service features. Document translation is Translator service.

Question 38: Which part of speech synthesis breaks text into words for pronunciation?

• A. Transcription
• B. Lemmatization
• C. Tokenization ✓
• D. Key phrase extraction

Rationale: Tokenization breaks text into individual words so each can be assigned phonetic sounds.

Question 39: Which TWO features enable text-to-text AND speech-to-text translation?

• A. Conversational Language Understanding
• B. Key phrase extraction
• C. Azure AI Speech ✓
• D. Azure Translator ✓

Rationale: Translator handles text-to-text; Speech service handles speech-to-text (which can then be translated).

Question 40: Which THREE sources can generate a knowledge base for Q&A?

• A. A webpage ✓
• B. An audio file
• C. An existing FAQ document ✓
• D. Manually entered data ✓

Rationale: Knowledge bases can be built from webpages, FAQ documents, and manual entries—but not directly from audio or image files.