A travel app needs to translate a user's spoken English into spoken Japanese and also display the Japanese text on screen. The app should minimize latency. What should the developer use?

Speech Translation with both text and speech output set to Japanese.

Speech-to-Text in Japanese and then translate the text to English.

Speech-to-Text in English, then Text-to-Speech in Japanese, and a separate translation API.

Custom Speech with a Japanese language model.

What is the purpose of Speech Synthesis Markup Language (SSML) in Azure Speech?

To translate speech into multiple languages

To recognize the speaker's identity

To control pronunciation, pitch, rate, and volume of synthesized speech

A voice-controlled kiosk in a museum must wake on 'Hey Museum' and then answer questions like 'What is the history of this painting?'. The kiosk has limited processing power. You need to minimize resource usage while still providing accurate answers. What should you do?

Use continuous intent recognition for both the wake word and the question.

Use keyword recognition for the wake word, then use intent recognition with LUIS to understand the question.

Use batch transcription to process all audio and then run intent recognition.

Use speech translation to translate the question into a database query.

A company is building a customer-facing IVR system using Azure Speech. The IVR must greet callers with a warm, natural voice, but the default neural voice sounds too formal. The team wants to adjust the speaking rate and add a short pause after the greeting without changing the text. What should they do?

Append extra spaces and punctuation to the input text to create pauses and slow the speech.

Use SSML with the and elements to control rate and pause.

Use the Speech Synthesis Markup Language element to set a custom style.

Select a different prebuilt neural voice that has a warmer tone and slower default rate.

A company is building a generative AI assistant that must respond with a specific brand voice and also adjust its speaking style based on the user's emotion (e.g., empathetic for complaints). The assistant uses Azure OpenAI for text generation and Azure Speech for audio output. What is the best way to achieve the brand voice and emotional style?

Use Custom Speech to train the model to speak with the brand voice.

Use Speech Translation to convert the text to the brand voice.

Use a Custom Neural Voice for the brand voice and SSML to adjust style.

Use a prebuilt neural voice and SSML to adjust pitch and rate.

Microsoft Certified:Azure AI Engineer Associate

Domain 5Objective 2

Process and Translate Speech AI-102 Practice Questions (Page 1)

Part of the Implement natural language processing solutions domain, which accounts for 15-20% of the AI-102 exam. Microsoft does not publish an official question count, but from its 100-minute exam (~40–65 total, ~6–13 in this domain), expect 2–4 from this objective — we provide 45 practice questions to prepare you well beyond it. (estimate)

45questions here

9free pages

9concepts

15-20%of the exam

Questions 1–5

1application · medium
A travel app needs to translate a user's spoken English into spoken Japanese and also display the Japanese text on screen. The app should minimize latency. What should the developer use?
Select an answer first
2foundation · easy
What is the purpose of Speech Synthesis Markup Language (SSML) in Azure Speech?
Select an answer first
3expert · hard
A voice-controlled kiosk in a museum must wake on 'Hey Museum' and then answer questions like 'What is the history of this painting?'. The kiosk has limited processing power. You need to minimize resource usage while still providing accurate answers. What should you do?
Select an answer first
4application · medium
A company is building a customer-facing IVR system using Azure Speech. The IVR must greet callers with a warm, natural voice, but the default neural voice sounds too formal. The team wants to adjust the speaking rate and add a short pause after the greeting without changing the text. What should they do?
Select an answer first
5expert · hard
A company is building a generative AI assistant that must respond with a specific brand voice and also adjust its speaking style based on the user's emotion (e.g., empathetic for complaints). The assistant uses Azure OpenAI for text generation and Azure Speech for audio output. What is the best way to achieve the brand voice and emotional style?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to AI-102

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by Microsoft. “AI-102” is a trademark of its owner, used for identification only.

Process and Translate Speech AI-102 Practice Questions (Page 1)

Questions 1–5

A travel app needs to translate a user's spoken English into spoken Japanese and also display the Japanese text on screen. The app should minimize latency. What should the developer use?

What is the purpose of Speech Synthesis Markup Language (SSML) in Azure Speech?

A voice-controlled kiosk in a museum must wake on 'Hey Museum' and then answer questions like 'What is the history of this painting?'. The kiosk has limited processing power. You need to minimize resource usage while still providing accurate answers. What should you do?

Keep studying — Implement natural language processing solutions