Keyword Spotting and Wake Words Quiz

Answers and explanations appear as you take the quiz.

1. What is the primary job of a wake-word detector?
   • To detect when a specific trigger phrase has been spoken
   • To transcribe every word the user says into text
   • To translate speech into another language
   • To compress audio for cloud storage
2. Why do many smart speakers use a two-stage detection design?
   • To support two different languages at once
   • To let a low-power chip do a rough check before a larger model confirms, saving battery
   • To record two copies of the audio for backup
   • To run two separate voice assistants simultaneously
3. What does a 'false accept' mean in wake-word detection?
   • The device ignores a correctly spoken trigger phrase
   • The device mishears one word as another in a sentence
   • The device wakes up even though nobody said the trigger phrase
   • The device fails to connect to the internet
4. Roughly how is the incoming audio prepared before the neural network sees it?
   • It is saved as a full-length MP3 file first
   • It is converted directly into written sentences
   • It is sent unchanged to a remote server for analysis
   • It is sliced into short overlapping frames and turned into acoustic features like MFCCs or mel energies
5. What trade-off can raising a wake-word detector’s activation threshold create?
   • Fewer false accepts but potentially more false rejects or delayed detections
   • More false accepts and guaranteed zero false rejects
   • No change in either error rate under any audio condition
   • A guarantee that every user’s phrase is recognized
6. What result did Hello Edge report for its depthwise separable CNN on the Google Speech Commands evaluation?
   • 100% accuracy on all languages and devices
   • 95.4% accuracy, about 10% higher than a DNN with a similar parameter count
   • 95.4% lower memory use than every competing model
   • A 10% guaranteed battery-life increase for every microcontroller
7. What does a false reject mean in wake-word detection?
   • The detector activates without the wake word
   • The microphone records silence correctly
   • The user says the wake word but the detector does not activate
   • A speaker is assigned the wrong identity
8. Why may a wake-word system use a compact model on-device?
   • It guarantees zero false accepts
   • It replaces all speech recognition
   • It makes the wake phrase unnecessary
   • It can reduce dependence on network access and support low-power always-on detection