Quiz 5 Flashcards
Consonants : Frics | Speech Perception | Perception Designs
T or F : all languages are rich in fricatives
F - english is, many are not
Low pitched fricatives vs high pitch (give examples of each)
- Low : [f, v, θ, ð, h]
- High : [s, z, ʃ, ʒ]
Frequency for FEMALES for 5 formants (range)
5 formants under 5500Hz
T or F : a LOW pitch fricative corresponds to a SIBILANT
F, they are HIGH pitch
Frequency for MALES for 5 formants (range)
5 formants under 5000Hz
Sibilants have an energy concentration in _____ frequencies
higher
Non-sibilants have an energy concentration in _____ frequencies
lower; across all
Which of the following will be darker on the spectrogram and WHY?
a. [s]
b. [z]
c. [t]
d. [f]
e. [v]
[s] and [f] because they are voiceless fricatives - which are DARKER (dispend more energy)
Which sound will have greater amount of energy in the higher frequencies?
A. [v]
B. [z]
C. both will be very similar
D. neither
[z] because it’s a sibilant (more turbulence)
What to measure for fricatives? And what is the measure USED?
Center of gravity; Hz
What does the center of gravity for frics tell us?
The average HEIGHT of the frequencies (for frics)
For fricatives, what is the typical window range that we will use in PRAAT to see the spectrum
up to 10 000Hz!
What is the normal/default frequency and sampling rate in PRAAT?
22050Hz, 44100Hz
The basilar membrane’s is in the _____ and its role is
inner membrane
processes sound vibrations at different points along itself, where it processes different frequencies
For vowels, what frequencies do we usually care about?
F1 and f2 - so 1000 to 2000hz ish
Men’s vocal folds have more ____ and vocal fold ______ which means that their vibrations will be slower
mass; thickness
Basilar membrane - 2 ends and what frequencies they process
- Base - THICK end : higher frequencies 5000Hz to 20kHz?
- Apex - THIN end : lower frequencies
T or F : the BASE end houses more frequencies
False, it is the APEX, meaning that we can distinguish low sounds way better than high ones
Tonotopic organization
diff parts of the membrane process diff. frequencies (tone)
For F1 frequency in the vowel [a], which end of the basilar membrane will be excited more?
A. Apex
B. Base
A. Apex
Which end of the basilar membrane will be active in processing the acoustic signals related to the sounds [s] and [z]?
A. Apex
B. Base
B. Base
Categorical perception
Refers to how our brains, depending on our language, will store/categorize particular sounds (some that can even contrast in another language)
- slight ‘fuzziness’ in between sounds
In categorical perception, looking at + and - VOT helps us categorize ______
voiced and voiceless stops perceived
For the perception of STOP place of articulation, we should look at :n(hint velar pinch)
A. F0
B. F1
C. F2
D. F3
C. F2
The speech perception study with /b/, /d/ and /g/ showed what results?
That between the sounds at CERTAIN intervals, there are both some very stable “i know it” regions and others that are sensitive/unsure of how to differentiate
In speech perception, what would sharply defined perceptual shifts indicate?
- That people were really sure of their answer.
- That the voicing remained constant for all 3, i.e. no one can actually distinguish the sounds
- That boundaries between sounds have been applied.
- That boundaries between sounds have been applied.
John is looking to find a perception experiment design that allows him to use 2 or more stimuli and whether or not they contrast eachother, without wanting his participants to know any categories. Should he choose :
A. Discrimination task
B. Identification task
A. Discrimination task
What are the 3 (2 + 1 e.g) defining parts of a DISCRIMINATION and IDENTIFICATION design
D
1. Looking at perceiving if 2+ simuli are seem the same/different
2. NO need to label/categorize the stimuli
3. Examples : AX, ABX, 2AFC
I
1. Label/categorize a stimulus
2. PREdefined categories create the name for the perceived stimuli
3. Examples : labelling, Yes-No tasks
T or F : An AX design will ask you to label a stimulus during each trial (2 trials total)
F - is discrim. + 2 sitmuli presented in each trial
In an AX design, why is it important to counter-balance the order of pairings (e.g. same, diff., same, diff.)
Prevents bias and misdirection + increases one’s attempt at accuracy
Difference between AX and ABX design
AX has two stimuli - 1 sound plus ? sound
ABX has 3 stimuli - 2 sounds + 1 ? sound
Main task of an ABX design
determine if X sound is the SAME as A OR B (not just similar!)
In Discriminative design tasks, what will an experimenter do to create an X sound
they will slightly manipulate the A or B sound, but one will always match the X sound presented
What is an ISI in discriminative tasks? what # measure does it usually take?
Interstimulus interval : it is the pause between one stimulus and the next - usually 500ms
T or F : an AX and an ABX task will both have 2 ISIs
F - just ABX
2AFC : discriminative or indicative?
discriminative
Explain a 2AFC task (Hint : beet vs bit) - the name + the instruction
Two Alternative Forced Choice Design
1. 2 stimuli presented for each trial
2. Person has to determine the order “Which stimulus came first, A or B” - trying to find [i]
Sound A : beet [i]
Sound : bit [I]
2AFC design elements (3) (Hint : BAD)
- (-) BIAS - good for close stimuli
- Assumes that EITHER order is possible
- “discernment’ of proper order
Main difference between AX + ABX AND 2AFC
First 2 look at same sound, 2AFC looks at the order
4IAX design - explain (think sounds, 1st and 4th, UNiq.)
4Interval Forced Choice Design
1. 4 sounds, 3 of the SAME - have to either choose the one that sounds ‘odd’ or the one that is same/different from the others
2. 1st and 4th sounds will never be the stimulus to think about - just 2 or 3
3. Look at the order/uniqueness of stimuli
Goal of a ‘yes/no/’ design
To test a person sensitivity to a specific segment of sound
T or F : a Yes/no design only has 1 stimulus presented for each trial
T - all the person has to do is say “yes i heard it “ or “no”
To make a Yes/No Design trickier/testable, what do researcehers do?
They manipulate the targeted sound a little bit to sound different
T or F : in a labeling design, we will manipulate the frequency of a fricative
F - the CENTER OF GRAVITY
What is the goal of the labeling design? How?
to assess one’s categorical knowledge + memory of the previous stimuli
They have to hear the stimulus and categorize it ‘is the ‘s’ you heard like in ‘seat’’?
Describe the oddity design - is it limited to # of stimuli? whats the task?
- Many stimuli, only 1 unique
- Take unique one OUT
- Often limits 3-4 stimuli
What is the measure of sensitivity in perception experiments?
d’ - d prime
What does ‘d’’ tell us in perception experiments?
Shows us how well a person can discern between a SIGNAL (e.g. stim) and NOISE (not correct/silence)
Difference between signal and stimulus
Signal looks at a very specific KIND of stimulus that detects/discerns between responses, usually talking about that vs background noise
Stimulus is anything really, it’s very general
d’ represents what in terms of DISTANCE
distance between the MEANS of the signal and noise distribution - uses SD of the noise
d’ = Z what?
Z(hit rate) - (False alarm rate)
Perception experiment : HIT is similar to ____ and Correct rejection is similar to _____
Miss; false alarm
HIT = ?
MISS = ?
FALSE ALARM = ?
COrrect REJECTION =?
Response and stimuli are diff - YES
Response is same and stimuli is diff - NO
Response is diff and stimuli is same - NO
Response is same and stimuli is same - YES
3 good pros about perception expeirments
- Allows for a bias-free measure of perceptual sensitivity.
- Commonly used in signal detection theory to quantify the ability to differentiate
between informative signals in the presence of noise. - Provides insights into the listener’s/perceiver’s inherent ability, free from decision biases.
What does the ‘center of gravity’ measure in the context of fricative sounds? (One correct answer.)
A. The duration of the fricative sound.
B. The center of the place of articulation for the sound involved.
C. The average frequency at which energy is concentrated in a spectrum.
D. The loudness of a sound measured in decibels.
C. The average frequency at which energy is concentrated in a spectrum.
In a spectrogram analysis, how do voiceless fricatives typically appear compared to their voiced counterparts? (One correct answer.)
A. They have more energy and appear darker.
B. They are only visible in the lower frequencies.
C. They have less energy and appear lighter.
D. They have the same energy and appearance.
A. They have more energy and appear darker.
When analyzing fricative sounds, which of the following statements are true? (Two correct answers.)
A. The center of gravity measurement for fricatives is expressed in decibels (dB).
B. The fricatives [ʒ] and [z] are examples of voiced sibilant fricatives.
C. Fricatives like [f] and [v] typically have energy distributed across all frequencies, unlike [s] and [z].
D. Voiceless fricatives generally appear lighter in spectrograms due to lower energy levels compared to voiced fricatives.
B. The fricatives [ʒ] and [z] are examples of voiced sibilant fricatives.
C. Fricatives like [f] and [v] typically have energy distributed across all frequencies, unlike [s] and [z].
Which of the following statements are true regarding the basilar membrane in the human auditory system? (Two correct answers.)
A. The base of the basilar membrane is thin and responds primarily to high-frequency sounds.
B. The base of the basilar membrane is thick and responds primarily to low-frequency sounds.
C. The apex of the basilar membrane is thin and responds primarily to high-frequency sounds.
D. The apex of the basilar membrane is thick and responds primarily to low-frequency sounds.
B. The base of the basilar membrane is thick and responds primarily to low-frequency sounds.
D. The apex of the basilar membrane is thick and responds primarily to low-frequency sounds.
According to the concept of categorical perception in linguistics: (One correct answer.)
A. All languages share identical distributions and categories for mapping acoustic information.
B. Acoustic information is perceived as a continuous sequence without distinct categories.
C. Listeners map acoustic information onto linguistic categories specific to their native language.
D. Acoustic variations within a category are perceived as completely different sounds.
C. Listeners map acoustic information onto linguistic categories specific to their native language.
