W3: speech perception, production and errors

Question 1

What is speech?

Answer 1

A continuous stream of sound - no gaps between words yet we can still understand and link to meaning

Question 2

Explain phonemes in speech?

Answer 2

They sound different in different contexts

Can vary due to: Loudness, excitement and the sounds of the words around it

Question 3

What is the segmentation problem?

Answer 3

There are no clear boundaries between words - they blur and things can come out sounding very different
E.g. Isle of view - I love you
Mister Abbot - Mr Rabbit

Question 4

How do we solve the segmentation problem?

Answer 4

Possible word constraint: we like to segment speech so that it maps onto whole, possible words

Meaning constraints: we also prefer the mapping to make sense
E.g. aspirin says Diana says - as princess Diana says

Question 5

What is the invariance problem?

Answer 5

Phonemes are not always pronounced/perceived the same way, in every context

They vary according to: 
The surrounding sounds
Speech speed
Speaker accent 
Speech formality

Question 6

Explain assimilation of the invariance problem

Answer 6

Sounds take on some of their neighbors properties

- ‘o’ is not normally a nasal sound but nasalised (air comes out of your nose) when with nasal consonants (song, gone)

Question 7

Explain co-articulation effects of the invariance problem

Answer 7

Sounds can be produced more quickly/easily (modified to fit with the next sound along)

The listener gets a clue as to what sound is coming next

Co-articulating - producing together

E.g. Thompson (the p came from the noise made in between thom and son until it was eventually spelt like this)

Question 8

Explain allophones in the invariance problem

Answer 8

Phonemes that are pronounced slightly differently but do not contribute to differences in meaning

Question 9

Explain the categorical perception of phonemes

Answer 9

Despite many variations in phonemes, we only ever perceive speech sounds as one phoneme or the other

Question 10

How do we categorise stop consonants?

Answer 10

It depends on the voice onset time (VOT)
Some consonants you can say for as long as you like (mmmm, ssss) because there is still air coming out - but there are some where you completely stop the air like (p, b, d, t)

Question 11

When you produce a stop consonant, what is the VOT?

Answer 11

The time between the ‘burst’ which is when you let the air come out and when you start voicing (start to move your vocal cords)

Question 12

VOT timing is the only difference between perceiving..

Answer 12

p or b
t or d
k or g

The only difference between these sounds is when you start moving your vocal cords after you have released the air

Question 13

Explain context in understanding words

Answer 13

Single words alone are much harder to understand than if they were in a sentence - especially if there is a noisy background

The perception of words in speech is influenced by higher-level knowledge of semantics and syntax

Question 14

Top-down processing in the understanding of spoken words?

Answer 14

Our context and expectations help us to understand the words - helps to decode words

Question 15

Explain a study on how top-down processing helps to restore phonemes

Answer 15

Warren and Warren
- Presented coughed over the word to participants, but each time in a different sentence
E.g. It was found that the *eel was on the shoe
it was found that the *eel was on the orange
- People didn’t realise the phoneme was missing, they restored the phonemes in their mind - they thought they heard it

Question 16

What is the process you go through for every word you hear?

Answer 16

‘Can’ - have sensory input to your lexicon

Have a selection phase in your lexicon - ‘can’ is activated in your lexicon more than other similar words

Recognise the words

Lexical access - what does ‘can’ mean?

Then integrate it into sentence

Question 17

What are the 4 models of speech perception?

Answer 17

1. Template matching
2. Analysis by synthesis
3. Cohort model
4. TRACE model

Question 18

What is the template matching model? are there any critiques?

Answer 18

Every word we hear is stored as a template in our lexicon
- when we hear a word we match it to our mental template
= recognition

BUT too much variation in speech for this to be plausible (so much variance in pronunciations, even accents - different versions that would need to be stored)

Question 19

What is the analysis by synthesis model of speech perception? any strengths and weaknesses?

Answer 19

Motor theory
We interpret the speech we hear by matching it to how we would produce speech ourselves

It accounts for speaker differences unlike the template-matching model - if you hear someone say something different to us we would then just go through how we would pronounce it to understand

HOWEVER, no explanation as to how you turn articulated sounds into the heard target
If someone says something completely weird, and unexpected we can still understand it (driven by data, not hypothesis)

Question 20

Is there any evidence of motor processes in speech perception?

Answer 20

When listening to others speak, brain imaging shows that the motor cortex is activated

Question 21

What is the cohort model of speech perception?

Answer 21

When we hear speech we are setting up a cohort of possible words to decide what we heard
- items are eliminated from this cohort until there is only one word left which is assumed to be the word heard

Question 22

What are the 3 stages of the cohort model of speech perception?

Answer 22

Access stage: when you hear a word it activates a set of words (cohort)

Selection stage: one item chosen from the cohort

Integration stage: words syntactic and semantic properties are used to integrate the word into sentence

Question 23

What is the TRACE model of speech perception?

Answer 23

Connectionist model - interactive

Context of word and sentence can facilitate the perception of individual sounds

Processing occurs through excitatory and inhibitory connections between processing units called nodes

NODES:

• each node has a resting level and a threshold so when you perceive distinctive features, phonemes and whole words, nodes become activated
• If it gets above the threshold, it is considered for matching the input and may excite or inhibit other nodes

Question 24

Which of the speech recognition models focuses on matching speech signals to phonetic segments?

Answer 24

Analysis by synthesis - motor theory

Question 25

Which of the speech recognition models describe the recognition of auditory words?

Answer 25

Cohort model

TRACE model

Question 26

What is the subglottal system and what is its role?

Answer 26

Providing air for speech

Lungs and associated muscles
Trachea

Question 27

What is the role of the larynx in speech production?

Answer 27

Used for phonation, making sounds

Question 28

What is the role of the vocal tract in speech production?

Answer 28

The moving and non-moving parts determine how sounds are articulated

Question 29

Explain the production of vowels and consonants?

Answer 29

Vowels: don’t obstruct the passage of air
Consonants: do obstruct the passage of air

Question 30

What are pure vowels?

Answer 30

They have just one vowel sound

Bit, bat, bet, bomb, boot

Question 31

What are diphthongs?

Answer 31

Two vowels produces in a smooth glide that moves from one vowel to another
Bake, bike

Question 32

What 3 things can you vary to produce vowels?

Answer 32

Tongue height - raised or lowered

Tongue position - front or back (most involved part)

Lip position - rounded or unrounded

Question 33

Explain the phonation process

Answer 33

Vocal folds open and close while puffs of air flow through the oral cavity

Voiced/phonated sounds are produced by this vocal fold vibration and unvoiced sounds are pronounced without it

Question 34

Speech errors are..

Answer 34

Quite common

Question 35

What are some types of speech errors? (2)

Answer 35

Spoonerisms: swapping word beginnings

Freudian slips: errors showing true thoughts

Question 36

Why are speech errors interesting for linguistics?

Answer 36

Tells us how speech is represented in the brain

If you can make an error on it, its represented in the brain

Question 37

How do you collect errors for research?

Answer 37

Recorded corpora (large amounts of recorded speech) 
record errors as they occur naturally e.g. radio audio BUT you have no control over what may be causing the errors 

Experiments
researcher tries to elicit speech errors by making task difficult or stressful BUT tasks may be too artificial

Question 38

What errors do we make with phonetic segments?

Answer 38

Voicing and nasality

Question 39

What phoneme errors do we make?

Answer 39

Anticipation errors: Later sounds come earlier than intended

Perservation errors: sounds produced earlier reappear later on

Exchange errors: units of varying sizes change places (your shrine spinks)

Question 40

Explain Garrett’s model of speech production

Answer 40

Conceptualisation: pre-verbal message

Formulation: translate it into linguistic form (words you need, what order they need to go in:syntax, sounds)

Execute: detailed phonetic and articulatory planning - articulation (say it)

Question 41

What does garrett’s model suggest are the two stages of syntactic planning

Answer 41

Functional level:

• content words are selected
• specify meaning
• words assigned to syntactic roles

Positional level:

• function words are selected (a, the)
• words are put in the correct order

Question 42

Sentence frames?

Answer 42

We specify a syntactic frame for our sentence - with slots for content words

Question 43

What provides evidence towards the two levels of syntactic planning (functional and positional levels)?

Answer 43

Words only get exchanged if they’re involved at the same level of processing

• Content and function words never change places
• words can change across phrases but always keep their syntactic class (noun remains a noun)
• Sounds only change places over small distances

Question 44

What shows that speech production isn’t serial as Garrett said?

Answer 44

Word-blend errors
2 words simultaneously retrieved from the lexicon - suggest happening in parallel
E.g. I expect - I suppose= I expose

Phrase-blend errors
Whole phrases seem simultaneously activated and cross over where they seem the most alike
‘I am making a cup of tea’ & ‘I am putting the kettle on’
= ‘I am making the kettle on’
Have activated and planned to and the ‘ing’ matches and causes a switch

Non-plan internal errors
Talking about something but thinking about something else at the same time and the thoughts intrude on the words and changes what you say

Environmental contamination
Changing individual sound in what you are saying
Trying to say: ‘it has been raining for a while’ but as you say this, someone runs towards you and you say ‘it has been running for a while’