Plan and manage an Azure AI solution →hardMultiple ChoiceObjective-mapped

AI-102 Plan and manage an Azure AI solution Practice Question

You deployed a custom neural voice in Azure AI Speech. The model generates poor prosody for long sentences. What should you do?

⚠ Common exam trap

A common mix-up: candidates think runtime SSML tags (Option D) can fix poor prosody, but the question specifies the model generates poor prosody, indicating a training data quality issue that SSML cannot correct.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

Re-record the training audio at 24 kHz sample rate.

A custom neural voice model requires high-quality training data. A 24 kHz sample rate is the standard for Azure AI Speech's neural voices, as it captures the full frequency range needed for natural prosody. Re-recording at this rate ensures the model learns proper intonation and rhythm for long sentences, directly addressing the poor prosody issue.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✓
Re-record the training audio at 24 kHz sample rate.
Why this is correct
High-quality audio is essential for learning natural prosody.
✗
Switch to a prebuilt neural voice.
Why it's wrong here
Prebuilt voices may not match the desired custom characteristics.
✗
Increase the number of training scripts.
Why it's wrong here
More scripts alone don't improve prosody if audio quality is insufficient.
✗
Use SSML tags to control prosody at runtime.
Why it's wrong here
SSML can adjust but cannot fix underlying model deficiencies.

Go deeper

AI-102 Plan and manage an Azure AI solution Practice Question

Why each option matters

Option-by-option breakdown

Related to this question