Speech and Hearing Science Exam 2 Study Guide

Sound

A physical phenomenon that describes a movement or vibration of an elastic medium without permanent displacement of the particles

Elasticity

tendency of an objects resistance to deformity and its return to the rest position

Inertia

tendency to resist change in motion (the greater the objects mass, the greater the inertia)

Brownian Motion

Roughly equivalent distance between molecules (no stored energy)

A force is needed to start air particle movement

True

1 multiple choice option

Recoil and inertia make particles vibrate

True

1 multiple choice option

When molecules are close together, pressure is very high

True

1 multiple choice option

Presure

a force exerted over a unit of time (proportional to density)

Atmospheric Pressure

Pressure of air molecules at rest

Compression

areas of high density and pressure

Rarefaction

thinning of air molecules that creates areas of decreased air pressure and density

Air molecules move around a rest position, they do not move through space

True

1 multiple choice option

A pressure wave moves through space

True

1 multiple choice option

Plane waves

sound moving away from the source (travel in straight lines)

Waveform axes

x-axis: time

y-axis: pressure

Period (T)

Amount of time needed to complete one cycle of vibration, measured in units of time (the time it takes for a vibrating object to return to its starting position and begin its repeating motion)

Frequency (f)

the number of complete vibratory cycles per unit time, measured in Hz

Spatial Measures

how much distance there is between 2 peaks of a single cycle

Wavelength (spatial measure)

Distance covered by a high-pressure region and succeeding low-pressure region

The higher the frequency, the lower the wavelength

True

1 multiple choice option

Short wavelengths strike objects whereas longer wavelengths require sound to bend around objects

True

1 multiple choice option

Sine waves

simple harmonic motions (ex: sitting on a swing)

(periodic, single frequencies that are the building blocks of complex waveforms)

3 parameters of sine waves (sinusoidal motion):

Frequency, amplitude, phase

Periodic sound

a sound that always has the same period

Waveform

Shows an acoustic event in a time domain

Spectrum

shows an acoustic event in frequency domain

Fourier Analysis

breaks down complex waveforms into simpler sine waves, allowing the analysis of a signal's underlying frequency components

Complex Periodic Waveform

sum of individual sinusoids at the harmonic frequencies - identifies the frequencies in a signal (fundamental frequency, harmonics, energy between) (every cycle looks almost identical to the one before it)

You do not have to have all harmonics in a signal

True

1 multiple choice option

Complex aperiodic events

no repetitive pattern and no harmonically related frequency components (frequency cannot be determined by waveform, but needs to be determined by spectrum)

Examples of periodic waveforms:

Simple (single sine waves) and complex (multiple sine waves)

Examples of aperiodic waveforms:

Continuous (noise) and Transient (pulse)

Resonance

object vibrates with maximum energy at a particular frequency (range), natural frequency

Vibratory energy can be transferred to objects

True

1 multiple choice option

Spring Mass Model (mechanical resonance)

period of vibration - amount of time required to complete one full cycle of motion (similar to the motion of air molecules)

Frequency of vibration is determined by mass and elasticity

True

1 multiple choice option

Increase in stiffness, increase the resonant frequency

True

1 multiple choice option

Increase in mass, decrease the resonant frequency

True

1 multiple choice option

Helmholtz Resonator

a vibrator consisting of a volume of enclosed air with an resonator open neck and resonator bowl (works the same as a spring mass model - has a single resonant frequency) (applicable to vowels)

If you put energy into a Helmholtz resonator, you will get a pure tone

True

1 multiple choice option

Helmholtz Resonator Neck

air within the neck of the Helmholtz resonator acts as a plug of air with mass (if mass increases, so does the resonant frequency) (Acoustic mass, Ma, can be increased by lengthening the neck or decreasing neck opening)

Helmholtz Resonator Bowl

Force applied to the bowl of a Helmholtz resonator acts as a spring ((If acoustic compliance, Ca, increases (less stiff), resonant frequency decreases))

Acoustic Mass

Ma - inertance (similar to inertia of mechanical masses)

Acoustic Compliance

Ca - inverse of stiffness

The larger the mass, the lower the resonant frequency

True

1 multiple choice option

The stiffer a system, the higher the resonant frequency

True

1 multiple choice option

Standing Waves

A pattern of vibration that simulates a wave standing still (vibrating air molecules produce the same pressure variation at the same location) (pressures within the tube appear "frozen" - peaks and nodes are always in the same spot)

Tube Resonators

Tubes of uniform cross-sectional area with both ends open

Any wavelength that has atmospheric pressure at its open ends is a resonant frequency of a tube opened at both ends

True

1 multiple choice option

If you put energy into a tube resonator, you will get a periodic sound

True

1 multiple choice option

Tubes with one end closed

Atmospheric pressure on the open end and greatest pressure along the wavelength on the closed end (if sending different frequencies through the tube, certain ones will produce strong pressure changes) (sound bounces back to the open end)

Spring mass-models and Helmholtz resonators only have one resonant frequency

True

1 multiple choice option

Damping

energy loss in vibratory systems (lightly damped systems or heavily damped systems)

4 factors causing damping:

Friction, absorption, radiation, gravity

Damping factors produce energy at frequencies other than resonant frequency

True

1 multiple choice option

Bandwidth

the maximum amount of data that can be sent in a fixed amount of time, usually measured in bits per second (index of tuning, range of frequencies between 3dB down point on either side of peak energy)

The damper something is, the wider the bandwidth of the spectrum

True

1 multiple choice option

Source

Input signal generated by vibrating vocal folds

Glottal Area Function

shows a graph of the change in glottal area as a function of time during voice production (durations between 1ms - 5ms - 8ms)

Inverse Filtering

look at the "source" of a sound even if we only have a recording of the speech

Characteristics of a signal

- Periodic

- Slope of the opening phase is shallower than that of the closing phase

- Shows time where the vocal folds are open (60%) and apart (40%)

- Complex periodic waveform, not a single sinus sound

Glottal Source Spectrum

Sound produced by the vibrating vocal folds ("source" spectrum is "filtered" by the vocal tract due to resonance)

Complex periodic event

waveform pattern repeats over time (non-sinusoidal shape)

Quasi-periodic

very small variations in successive glottal cycles (voice is this)

"Tilt" of a waveform

the steeper the closing slope, the less tilted the spectrum

Hyperfunctional voices

very rapid closing phase, less than normal tilt, "pressed voice" (cheerleaders, overuse, nodules)

Hypofunctional voices

Slow closing phase, more than normal tilt, weak/breathy voice

Less open time = less tilted spectrum

True

1 multiple choice option

Filter

During vibration, the vocal folds snap shut, airflow is blocked, and air above the vocal folds is compressed/starts a pressure wave (vocal tract as tube closed on one end, no airflow at the vocal folds - airflow at the lips)

Resonance does not change F0 or harmonics

True

1 multiple choice option

First 3 Formants most important for speech:

- Formant 1: 560 Hz

- Formant 2: 1680 Hz

- Formant 3: 2800 Hz

Area function of the vocal tract

a plot of cross-sectional area as a function of distance along the vocal tract from the glottis to the lips

Source + Filter

As the sound wave flows through the vocal tract, some frequencies are amplified (formants) and others dampened (vocal tract shapes the input signal)

Formants

regions of resonance in the vocal tract (horizontal dark bands present for vowels, diphthongs, semi vowels, and nasals) (easy to visualize)

Spectral Envelopes

created by linear predictive code analysis (individual harmonics are less important, so they are often depicted as this - line on top of the curve)

Formant Bandwidths

factors responsible for energy loss are active in the vocal tract (nasal sounds = low intensity)

Absorption

when a frequency is close to the resonant frequency of a surround structure, energy is absorbed which increases damping

Radiation

loss of energy going from an enclosed tube to the open environment

Narrow bandwidths

Vowel resonances are fairly "sharply tuned"

Resonant Frequencies in a constricted tube

Resonance of tube closed at one end - Constriction located at pressure maximum (raises resonant frequency, constriction increase stiffness of air molecules)

The greater the constriction, the greater the increase in resonant frequency

True

1 multiple choice option

Constrictions in tube with multiple frequencies

1. any construction affects affects all resonant frequencies

2. a given resonance may be affected by 2 simultaneous constrictions

perturbation theory

if you change the constriction/shape of the vocal tract, it changes the resonant frequency

Every single harmonic is impacted by a constriction

True

1 multiple choice option

3-parameter model of Stevens and House:

Tongue height, tongue advancement, lip rounding

Tongue height

F1 varies inversely with tongue height (the higher the tongue, the lower the F1) (more pronounced for front vowels vs back vowels)

Tongue advancement

F2 increases and F1 decreases with increasing tongue advancement (F2 increases the constriction to the front)

Lip rounding

all formant frequencies decrease with increased round of the lips (great decrease in F2, less effect on F1 and F3) (the higher the tongue, the larger the decrease)

3rd formant is not easily related to changes in articulatory dimensions

True

1 multiple choice option

Different vocal tract configurations can produce the same f-pattern

True

1 multiple choice option

Resonant Frequency

a frequency that a system likes to vibrate at

F1

tongue height

F2

tongue advancement

Spectrogram

Visual representation of sound frequencies over time

Analog experiments

Electrical components stimulated the vocal tracts (stevens & house developed the whole theory based on an electrical model)

Hunan Experiments

comparing calculated formant patterns from area functions with people producing the same vowels (if the calculated and actual formants match, theory is confirmed)

Vowels are always voiced, continuous and made in an open vocal tract

True

1 multiple choice option

Consonants have constriction in the vocal tract

True

1 multiple choice option

Nasal sounds involve 2 tubes whereas vowels involve only a single tube

True

1 multiple choice option

Glides/liquids are semi-vowels

True

1 multiple choice option