Speech recognition and voice quiz

Question 1

What is speech recognition?

Accepted Answer

Turning spoken words into text a computer can use

Answer

Making a computer speak out loud

Answer

Translating text from one language to another

Answer

Recognizing faces in photos

Question 2

What does the abbreviation ASR stand for?

Accepted Answer

Automatic Speech Recognition

Answer

Audio Sound Recorder

Answer

Advanced Sentence Reader

Answer

Artificial Speaking Robot

Question 3

What is the difference between speech recognition and text-to-speech?

Accepted Answer

Speech recognition turns voice into text, TTS turns text into voice

Answer

Speech recognition makes voice from text, TTS does the reverse

Answer

They are exactly the same thing

Answer

Both only work with music

Question 4

Why does a speech recognition system need lots of example recordings to learn?

Accepted Answer

To learn sound patterns from many voices and accents

Answer

So the program runs faster

Answer

Because microphones require it

Answer

To save battery power

Question 5

What is a microphone's job in a voice system?

Accepted Answer

It converts sound waves into an electrical signal

Answer

It plays music to the user

Answer

It translates languages

Answer

It stores passwords

Question 6

What does it mean to 'sample' an audio signal when a computer records sound?

Accepted Answer

Measuring the sound wave many times per second as numbers

Answer

Tasting the sound

Answer

Deleting parts of the sound

Answer

Making the sound louder

Question 7

Why do background noises like traffic or music make speech recognition harder?

Accepted Answer

Noise mixes with the voice and confuses the computer

Answer

Noise makes the microphone hotter

Answer

Noise makes words louder and clearer

Answer

Noise has no effect at all

Question 8

What is a 'wake word' in a voice assistant?

Accepted Answer

A phrase that tells the assistant to start listening

Answer

A word that turns off the internet

Answer

A password for your email

Answer

An alarm clock sound

Question 9

What is an accent, and why can it affect speech recognition accuracy?

Accepted Answer

A way of pronouncing words that systems must learn to handle

Answer

An accent is a type of microphone

Answer

A computer error message

Answer

A kind of background music

Question 10

What is a transcript in the context of speech recognition?

Accepted Answer

The written text of what was spoken

Answer

A picture of a sound wave

Answer

A type of microphone

Answer

A list of phone contacts

Question 11

How do live captions on videos use speech recognition?

Accepted Answer

They turn spoken audio into on-screen text in real time

Answer

They change the video colors

Answer

They make the video louder

Answer

They block advertisements

Question 12

What is a 'language model' in a speech recognition system?

Accepted Answer

A part that knows which word sequences are likely

Answer

A model of the human mouth

Answer

A list of foreign countries

Answer

A type of microphone

Question 13

Why might a speech system confuse the words 'two', 'to', and 'too'?

Accepted Answer

They sound the same but are spelled differently (homophones)

Answer

They are spoken at different volumes

Answer

They are in different languages

Answer

They are never used together

Question 14

What is voice recognition (speaker identification), and how is it different from speech recognition?

Accepted Answer

Voice recognition identifies who is speaking; speech recognition identifies the words

Answer

They are identical

Answer

Voice recognition only works on music

Answer

Speech recognition identifies the speaker's age

Question 15

Give one helpful use of speech recognition for people with disabilities.

Accepted Answer

It lets people control computers or get captions hands-free

Answer

It makes phones more expensive

Answer

It only helps gamers

Answer

It deletes old files

Question 16

What is a 'word error rate' (WER) used to measure?

Accepted Answer

How many mistakes the transcript has compared to correct words

Answer

How fast the computer runs

Answer

How loud the speaker is

Answer

How many languages are supported

Question 17

Why should voice assistants ask permission before recording your voice?

Accepted Answer

Because your voice is personal data and privacy must be respected

Answer

Because recording is illegal everywhere

Answer

Because it makes the device faster

Answer

There is no reason to ask

Question 18

What does 'real-time' mean for a speech recognition system?

Accepted Answer

Converting speech to text almost instantly as you speak

Answer

Working only at night

Answer

Recording for exactly one hour

Answer

Working without electricity

Question 19

What is a phoneme?

Accepted Answer

A small basic unit of sound in speech

Answer

A whole sentence

Answer

A type of phone

Answer

A computer chip

Question 20

How can a neural network help with speech recognition?

Accepted Answer

It learns sound patterns from many examples to predict words

Answer

It only stores files

Answer

It makes the microphone louder

Answer

It blocks the internet

Question 21

Why do speech systems sometimes struggle with children's voices?

Accepted Answer

Their higher pitch and pronunciation differ from adult training data

Answer

Children speak too quietly to record

Answer

Children use special microphones

Answer

The system dislikes children

Question 22

What is a spectrogram and why is it useful for speech recognition?

Accepted Answer

A picture showing how sound frequencies change over time

Answer

A list of words

Answer

A type of speaker

Answer

A volume button

Question 23

What is the risk of a 'deepfake voice'?

Accepted Answer

An AI-faked voice that can trick or deceive people

Answer

A voice that is always too quiet

Answer

A normal phone recording

Answer

A type of microphone

Question 24

Why can speaking too fast or mumbling reduce recognition accuracy?

Accepted Answer

Blurred sounds give the system fewer clear clues

Answer

Fast speech damages the microphone

Answer

It uses more battery

Answer

It changes the language

Question 25

What does it mean for a speech model to work 'offline' versus 'online'?

Accepted Answer

Offline runs on your device; online sends audio to a server

Answer

Offline means broken; online means working

Answer

They mean the same thing

Answer

Offline only works at night

Question 26

How do voice assistants understand a command like 'set a timer for 5 minutes'?

Accepted Answer

Speech-to-text, then understanding the intent, then acting

Answer

It guesses randomly

Answer

It calls a human helper

Answer

It only beeps

Question 27

Why is having diverse training data important for fair speech recognition?

Accepted Answer

It reduces bias so the system works fairly for everyone

Answer

It makes the app cheaper

Answer

It is required by microphones

Answer

It only helps adults

Question 28

In simple terms, what are 'features' that a speech model uses from audio?

Accepted Answer

Useful measurements pulled from the sound

Answer

Decorations on the screen

Answer

The price of the app

Answer

The name of the speaker

Question 29

What is one limitation of speech recognition you should keep in mind?

Accepted Answer

It can mishear words and should be double-checked

Answer

It is always 100% perfect

Answer

It only works for robots

Answer

It never makes mistakes with names

Question 30

How does speech recognition help when translating spoken languages?

Accepted Answer

It captures the spoken words as text for translation

Answer

It plays music between languages

Answer

It deletes the original language

Answer

It only works for written letters

Test yourself