speech science exam 2

the graph which displays amplitude over frequency and enables us to view individual formant frequency "peaks" is a(n)

spectrum

T/F: F1 and the first harmonic are directly related and often the same value

false

Resonant (formant) frequencies, like harmonics, are measured in

hertz

T/F: The resonant (formant) frequencies of the vocal tract are directly related to the fundamental frequency of the vocal folds.

False

In a general sense, F1 is related to the ___________ cavity. F2 is related to the ____________ cavity.

pharyngeal; oral

refers to the resonant frequencies of the vocal tract

formant frequencies

Refers to the vibrating frequency of the vocal folds

fundamental frequency

According to the Quarter Wave Resonator Theory, an individual with a vocal tract length of 17.5 cm who breathes sulfur hexafluoride (with a speed of sound of 133 meters/second) will have approximately which of the following formant frequencies when producing the schwa vowel?

190 Hz; 570 Hz; 950 Hz

compared to other vowels, /i/ is produced with the tongue in the highest, most anterior position in the vocal tract. For this reason it also yields a ____________ when acoustic measures are obtained.

high F2 and low F1

when a resonating cavity is larger, the resonating frequency will be ____________. when a resonating cavity is smaller, the resonating frequency will be ____________.

lower; higher

individual harmonics

narrowband

formant frequencies

both narrowband and wideband

glottal pulses

wideband

What are examples of temporal acoustic measures?

word duration, vot, speech rate

T/F: for the most accurate assessment of speech rate in oral readings, narratives, or conversation, one should measure in words per minute.

False

T/F: a stop watch is sufficient for measuring the duration of single syllables and words.

False

when measuring the VOT of /g/ in the word /gus/, one would expect a value that is

< 30 ms

on which type of spectrogram are formant frequencies easier to visualize?

wideband

which method would we consider the fastest and most accurate for extracting formant frequencies from a speech sample?

linear predictive coding

You run a Multi-Dimentional Voice Profile on a middle aged female. Her results indicate: jitter of 0.8%, shimmer of 3.22%, average fundamental frequency of 126 Hz. Based on these results, you would conclude that:

Her F0 is below average but jitter and shimmer are normal

Why are F1 and F2 values so significant for vowels?

they provide information on appropriate tongue position in the oral cavity during vowel production; in some cases they can provide information on place of articulation; the difference between F1 and F2 can specify the type of vowel produced

What are the predicted first three formant frequencies of the shwa volume (for an adult male)?

500 Hz, 1500 Hz, 2500 Hz

If the F2 of an adult producing the schwa vowel is 2100 Hz, then what is the value of F1?

700 Hz

The F1 is accounted for by the resonance of which cavity of the vocal tract?

pharyngeal cavity

The F2 results from the resonance of which cavity?

oral cavity

In terms of vocal tract configuration, why the /i/ vowel has such a large difference between F1 and F2

The vowel is produced with a small oral cavity and a large pharyngeal cavity

What is considered multiples of the fundamental frequencies

Harmonics

What do the thick, horizontal lines represent?

Formant frequencies

Where is the steady state part of the formant?

Vowels

Where is the formant transition?

Consonants

Is the formant transition always at the beginning of the word or syllable?

No, words that end in consonants also have formant transitions.

Which vowel has the greatest difference in value between F1 and F2

/i/

Which vowel has the least differences between F1 and F2

/a/

How many formants are necessary to recognize/synthesize a particular vowel

2 (F1 and F2)

How does rounding the lips affect the value of F2?

F2 will be lower because the oral cavity is getting larger; F1 will also become lower

When considering a spectrogram of a CV syllable, the transition portion of the formant will be associated with the __________ and the steady state portion will be associated with the __________.

consonant; vowel

Why is it considered more accurate to express speech rate in syllables-per-second instead of words-per-minute?

Words vary in length and number of syllables.

Why is measuring duration of speech samples using a stopwatch not accurate enough for research purposes?

Using a stopwatch factors in reaction time but in research time has to be exact to the millisecond.

Which are the temporal measures of speech?

vowel duration, voice onset time, rate of speech

Variable that have been shown to affect VOT are?

gender, native language, lung volume, motor speech impairment

What is the most basic frequency measure of speech?

fundament frequency; then, F1 and F2

Formant frequency contours are best visualized on which type of spectrogram?

wideband

The very thin, vertical lines visible only on wideband spectrograms represent

glottal pulses

The very thin, horizontal lines viewed only on narrowband spectrograms represent

individual harmonics

What is the name of the algorithm which breaks a complex waveform down into its simple sinusoid components?

fourier analysis

When producing a single vowel, formant frequencies (as displayed on a spectrogram) appear "steady state". This is because the

vocal tract remains in a relatively fixed position

What kinds of acoustic measures may indicate a voice disorder is present

when jitter, shimmer, fundamental frequency, amplitude are outside of the norm

T/F: Jitter should be above 1.04% in a "normal" voice.

False

Which of the following speech analysis software programs can be downloaded for free on your personal computer and utilized in both clinical and research settings?

Praat

What prosodic components may be altered to change the meaning or emotion in one's speech?

fundamental frequency, duration, amplitude

T/F: According to the breath group theory of intonation, declarative sentences are produced with a relatively steady F0 contour followed by a sudden drop off in the terminal portion.

True

According to the Functional Load Hypothesis, the neurological representation of prosody is as follows:

The left hemisphere generally processes linguistic prosody; The right hemisphere generally processes emotional prosody.

What is the term we use to refer to changes in fundamental frequency over connected speech?

Intonation

Would a jitter level of 4.6% be normal or indicative of a voice disorder?

indicative of a voice disorder

list examples of acoustic measures of speech

vowel duration, vot, rate of speech

list examples of frequency measures of speech

fundamental frequency, formant frequency, jitter

List 5 variables that would influence a patient's Voice Onset Time

gender, native language, lung volume, motor speech impairment, normal aging

explain the difference between Jitter and Shimmer

Jitter measures cycle to cycle frequency variation; shimmer measures cycle to cycle or period to period amplitude variation

For what kind of disorder would we most likely take measurements of jitter and shimmer?

voice disorders

Speech can be visually displayed as frequency across time on a ___________________, which is created by a machine called a ___________________.

spectrogram; spectrograph

Regarding formant transitions, the steady-state portion of the formant is related to the ____________________ portion of a syllable, while the transition portion of the formant is related to the ____________________ portion of a syllable.

vowel; consonant

What is voice onset time?

Voice onset time is the amount of time between the release of the stop closure and the onset of voicing.

A voiceless stop consonant will have a (longer/shorter) voice onset time than a voiced stop consonant.

longer

Should the voiced or voiceless stop consonant have a longer VOT?

The voiceless consonant should have a longer VOT because vocal fold vibration is delayed after articulatory release and voiceless stops are aspirated.

Explain, in terms of vocal tract configuration, why there is usually a large difference between the F2s for /i/ and /ɑ/.

There is a large difference between the F2 of /i/ and the F2 of /ɑ/ because of the position of the tongue. When producing /i/, the tongue is higher and more forward in the mouth making the F2 higher. When producing /ɑ/, the tongue is lower and more backwards in the mouth making the F2 lower.

For vowels, we generally have only two resonating cavities instead of three. F1 is the resonant frequency associated with the __________________ cavity and F2 is the resonant frequency associated with the __________________ cavity.

pharyngeal; oral

Moving the tongue forward in the mouth will result in a (bigger/smaller) oral cavity and a (bigger/smaller) pharyngeal cavity.

smaller; bigger

Moving the tongue backward in the mouth will result in a (bigger/smaller) oral cavity and a (bigger/smaller) pharyngeal cavity.

bigger; smaller

If a smaller resonating cavity renders a higher resonating frequency, then would the F1 become higher or lower in value when the tongue moves forward in the mouth?

F1 would become lower

The melody of speech

prosody

3 acoustic features of prosody, their unit of measure, and their perceptual correlates

amplitude = dB = loudness
duration = ms = length of time
fundamental frequency = Hz = pitch

Two types of prosody

linguistic and emotional

difference between linguistic and emotional prosody

linguistic conveys meaning of word; emotional conveys message

How does the fundamental frequency at the beginning of a statement compare to the fundamental frequency at the end of a statement?

Fundamental frequency is higher at the beginning of a statement than it is at the end.

Why are questions generally produced in a shorter duration of time than statements?

Questions are produced in a shorter duration of time than statements because speakers are perserving lung volume for the rise in fundamental frequency at the end of the question.

Fo averages for men, women, and children

Men: 100 Hz
Women: 200 Hz
Children: 300 Hz

phonemic transcription

- / /
- Recording only the minimal contrasts in English
- Only the basic IPA symbols

phonetic transcription

- [ ]
- The fine details of the pronunciation are for phonetic transcription
- Used to transcribe something that has actually been spoken
- Often use diacritics
- differentiate allophones

front vowels

i, ɪ, e, ɛ, æ

back vowels

u, ʊ, o, ɔ, ɑ

central vowels

ə, ʌ, ɚ, ɝ, a

Source-Filter Theory of Vowel Production

- a theory of speech production whereby the vocal folds are considered the sound source and the vocal tract the filter for the sound
- source is phonation or the vibrating vocal folds
- filter is the particular shape of the vocal tract

Fourier analysis

breaks a complex waveform down into its simple sinusoid component

Benefit of two separate resonating cavities in the vocal tract

allows us to create acoustically stable speech sounds

Relationship between formant frequencies and vocal tract shape/tongue position

Formant frequency values correspond with the position of the tongue in the oral cavity.

vocal tract configuration for /i/

a small oral cavity and a large pharyngeal cavity

vocal tract configuration for /a/

a large oral cavity and a small pharyngeal cavity

The effect of lip rounding on formant frequencies

F2 will be lower because the oral cavity is getting larger; F1 will also become lower

Quarter Wave Resonator formula for calculating formants of a schwa vowel

ss/(4 x L) = F1
F1 x 3 = F2
F1 x 5 = F3

Number of formants necessary to create a vowel

3

temporal measures of speech

vowel duration, voice onset time, rate of speech

frequency measures of speech

fundamental frequency, formant frequency, jitter

Definition of VOT

the amount of time between the release of the stop closure and the onset of voicing

Variables that affect VOT

Gender; Ethnic background; Native Language; Lung volume; Word position in a sentence

VOT of voiced consonants

VOT of voiceless consonants

>30 ms

Formant frequency characteristics of /i/

largest difference between F1 and F2

Formant frequency characteristics of /a/

smallest difference between F1 and F2

Characteristics and uses of narrowband spectrograms

display harmonics; bands are horizontal

Characteristics and uses of wideband spectrograms

display formant frequencies; bands are vertical