11 - Acoustics 1

Question 1

Name one reason acoustics are important for speech

Answer 1

• can provide objective physical measure of speech disorders
• can i.d. specific parameters of disordered speech
• non-invasive, inexpensive, tech-friendly, readily available (e.g. Praat)
• can be used to make fairly reliable inferences about certain physiological events
• specific disordered acoustics parameters can be used for differential diagnosis, targets in treatment, biofeedback therapy, and treatment outcome measures

Question 2

As sound moves through air, it stimulates air particles to experience compression and ______

Answer 2

Rarefaction

Question 3

Is the velocity of particles greater at points of compression or rarefaction?

Answer 3

Rarefaction

Question 4

Wavelength = ______ / frequency

Answer 4

velocity of sound

Question 5

If a tone is presented at 200 Hz, what would the next 3 harmonics be?

Answer 5

400, 600, 800

Question 6

What is the roll off rate for speech?

Answer 6

12 dB per octave

Question 7

Output = source x radiation x ______

Answer 7

filter

Question 8

The output is a combination of harmonics and ________

Answer 8

Formants

Question 9

What two things do formants depend on?

Answer 9

Vocal tract length

Cross-sectional shape of vocal tract

Question 10

What are formants?

Answer 10

Resonance frequencies of the vocal tract

Question 11

Describe the “odd quarter wavelength relationship”

Answer 11

With a uniform tube closed at one end (like vocal tract being closed at vocal folds):

• “odd” refers to the formants being at odd multiples of the lowest resonance frequency
• “quarter wavelength” refers to the areas of maximum velocity (antinodes, best resonance)

Question 12

In the formula to find a formant:
F=((2n-1) x c)/4L
what do n, c, and L refer to?

Answer 12

n = formant number (e.g. first formant is 1)
c = speed of sound (34,400 cm/sec)
L = tube length (17 cm vocal tract)

Question 13

A node is a point of volume velocity ______ (minima/maxima)

Answer 13

Minima

Question 14

Perturbation Theory stipulates that constricting a tube at volume velocity maximum causes a(n) _______ (reduction/increase) in frequency of the related formant

Answer 14

Reduction

Question 15

What is the frequency of the most prominent nasal formant?

Answer 15

300 Hz (lower than 500 Hz for vocal tract because essentially lengthens vocal tract from 17 cm to 29 cm)

Question 16

How do you get turbulent noise?

Answer 16

Generate high velocity air through constriction of vocal tract, causing eddies

Question 17

__________ is the term for the critical flow velocity at which turbulence occurs (shift from laminar to turbulent flow)

Answer 17

Reynolds number

Question 18

What is the critical Re value for turbulent speech noise?

Answer 18

Re = 1800

Question 19

A ______ (smaller/larger) cross sectional area requires less flow velocity to achieve the critical Re

Answer 19

smaller

Question 20

How are turbulent fricative noises shaped (filtered)?

Answer 20

By the resonating properties of the oral cavity in front of the occlusion

Question 21

The Source Filter Model of speech acoustics includes what 3 sources?

Answer 21

• voicing
• turbulent noise
• transient bursts

Question 22

Name 1 of the 3 considerations related to vocal tract length?

Answer 22

• lip protrusion
• laryngeal lowering
• children’s vocal tracts

Question 23

How does nasal resonance influence the first formant?

Answer 23

Lower first formant (300 Hz) related to an increased length of the vocal tract

Question 24

How do damping effects relate to nasal resonance?

Answer 24

Formants tend to have lower amplitudes (related to the nasal tissues acting as sound absorbers)

Question 25

How does nasal resonance influence antiformants or zeros?

Answer 25

- nasals are produced with a bifurcation (splitting) of the vocal tract into 2 sections: - open nasal cavity - closed oral cavity - acts as sound absorber for specific frequencies (absorbed frequencies referred to as zeros/antiformants and appear as large valleys in amplitude/frequency spectra

Question 26

The further back the place of articulation, the ____ (lower/higher) the frequency of the nasal zero

Answer 26

higher

Question 27

Each of the 3 places of articulation for nasal resonance are associated with specific zero values. What are the 3 nasal sounds of articulation?

Answer 27

/m/: 750-1250 Hz /n/: 1450-2200 Hz 'ng': >3000 Hz

Question 28

Which types of phonemes are associated with turbulent noise?

Answer 28

Fricatives Affricates Stops

Question 29

How is turbulent noise generated?

Answer 29

- noise generated by air passing through narrow constriction - narrow constriction causes jet to form, which generates eddies (irregular rotations of air pressure and velocity) - these eddies are referred to as turbulence, which generates random frequencies of sounds or noise

Question 30

What is the formula for Re (value to determine turbulence)?

Answer 30

Re = (flow velocity/cross sectional area) / viscosity constant for air

Question 31

How is fricative resonance related to turbulent noise?

Answer 31

The turbulent fricative noise is shaped (filtered) by the resonating properties of the oral cavity in front of the occlusion e. g. for /s/ and 'sh' the resonator acts like a very short tube closed at one end - this short tube results in a very high resonance (first formant) frequency, thus each fricative has specific resonance frequencies

Question 32

How do resonance frequencies relate to posterior place of oral constriction (e.g. 's' vs 'sh')?

Answer 32

Resonance frequencies decrease with posterior place of oral constriction ``` Fn = (2n-1) * c / 4L where L=2 cm (/s/) F(1) = (2(1)-1) * 34,400 / 4*2 F(1) = 34,400 / 8 F(1) = 4300 Hz ```