Text to Speech Quiz

Answers and explanations appear as you take the quiz.

1. What does text-to-speech produce from written input?
   • Spoken audio
   • A translated transcript only
   • A list of speaker identities
   • A set of image labels
2. Why might “$12.50 on 03/04” need revision before narration?
   • It is always impossible to synthesize numbers
   • The currency and date may need explicit wording for the intended locale
   • The sentence has too many letters
   • Every TTS engine rejects punctuation
3. A voice reads a technical name incorrectly. Which check would help?
   • Increase the output file size
   • Assume the pronunciation is acceptable because the voice sounds natural
   • Listen to the actual output, including uncommon terms
   • Change the sample rate and skip listening
4. Which statement about synthesis controls follows the guide?
   • Every engine supports identical pronunciation controls
   • Markup always changes every voice in the same way
   • A pronunciation dictionary fixes all timing issues
   • Supported controls differ, so test the selected engine and voice
5. What should accompany a narrated article for accessible use?
   • A required background soundtrack
   • Playback controls and an accessible text equivalent
   • Only a transcript locked from copying
   • A faster voice with no pause option
6. Why should speech review check word accuracy as well as how natural the delivery sounds?
   • Pleasant delivery guarantees every word is correct
   • The two qualities cannot be reviewed separately
   • Audio can sound pleasant yet mispronounce words or emphasize them misleadingly
   • Delivery matters only for music
7. What is a sensible way to handle long narration?
   • Divide it into meaningful segments and allow resume or repeat
   • Remove all playback controls
   • Make one uninterrupted file with no text alternative
   • Convert every sentence into a separate unrelated clip
8. Before using a voice that could be mistaken for a real person, what should be reviewed?
   • Only the speaker volume
   • Whether the output is a stereo file
   • The number of words in the script
   • Voice permissions and disclosure for the intended use