Acoustics final ch. 4 and 5

• source characteristics
• the transfer function of the vocal tract
• the radiation characteristic as the sound moves from the enclosed vocal tract to the environment

tells you what will happen to certain frequencies- how energy is transferred from input to output

• start with a tube of:
• uniform cross-sectional diameter
• open at one end and closed at the other
• set length of tube to be appropriate for an adult male or female (male- 17 cm, female- 14.5 cm, child- 8.75 cm)
• the output will resemble the vowel "uh"
• this model is valid out to 5 kHz
• *long wavelengths act the same way in a straight line as they do in a curved tube

• a function of the length of the tube
• best excited at a frequency of 1/4wavelength (f = c/wavelength)
• odd multiples of this lowest resonant frequency
• changing with changes in the length (men vs women, adults vs children, Fn = (2n-1)c/4l)
• resonant frequency will occur at the odd quarter length formula
• /uh/ has least amount of constriction

• the mouth opening is a boundary
• higher frequencies are resonated with greater amplitude than lower frequencies (atmosphere offers greater impedance to lower frequencies which results in larger particle displacement)

• displacement is not affected as much
• newton's third law of motion (for every action, there is an equal and opposite reaction)

• **trick! (talking about atmosphere?)
• -12 dB/octave leaving the larynx
• +6 dB/octave leaving the lips
• -12 + 6 = -6 dB/octave in the atmosphere

• a wave that remains in constant position
• constructive interference

change in direction of the wave

positive and negative values are less positive or less negative than normal and lose amplitude

where the lips meet the atmosphere (a boundary)

at particular frequencies determined by the volume velocity (U) or pressure (P)

• at the closed end of the tube
• P is at a maximum

• at the open end of the tube
• P is at a minimum

at the open end of the tube model

at the closed end of the tube model

as they come up and out of the mouth

• vibration will be reinforced
• frequencies at which the wavelengths have a maximum velocity (or minimum pressure) at the tube opening (lips) will be transmitted most effectively (with the greatest relative amplitude)

• 1/4, 3/4, 5/4, etc times the length of the tube (quarter length multiples)- the resonances of the tube
• these are the only wavelengths with the appropriate max (min) pattern

antinodes

• nodes
• max amp of vibration at antinodes

• the vocal tract
• they exist whether they are energized or not

• they modify it
• they DO NOT add energy to the speech signal
• resonances do not make you louder

the transfer function- tell you what will happen to that frequency whether its there or not

slim

articulators

• change the frequency of the formant
• is a perturbation of the standing wave in the vocal tract

they are used interchangeably

• volume velocity is maximum
• pressure is minimum
• formant frequency will be lowered

• volume velocity is at a minimum
• pressure is maximum
• formant frequency is raised

• antinode- lowered by a constriction in the oral cavity near a volume velocity maximum (lips)
• node- raised by a constriction in the pharynx

• antinode- lowered by a constriction (at the lips or in the oropharynx)
• node- raised by a constriction (in the anterior oral cavity)
• *pg 31

• antinode- lowered by a constriction (at the lips or in the middle of the oral cavity)
• node- raised by a constriction (in the oropharynx or in the anterior oral cavity

labial constriction

a constriction near the larynx

cover range of articulations needed to produce all vowels in English

• tongue height
• tongue advancement
• lip rounding
• corner and point vowels

• they have a fair degree of independence, but articulator movements affect each other
• thus, formant frequencies are a product of the entire length and shape of the vocal tract

because it's not the absolute values of the frequency that help us to distinguish, but the relationship of the formant frequencies

• tense vowels are relatively long
• lax vowels are relatively short

• a phonetic description
• supposedly correlated with articulatory effort
• actually correlated with duration

• two vowels forming a single nucleus
• on-glide and off-glide
• should see movement in formants

high vowels have a relatively higher fundamental frequency

• when you raise your tongue, your larynx raises too (makes pitch higher)
• hypoglossus pulls larynx up for high vowels, up to a 20 Hz difference

in groups according to their significant acoustic and articulatory properties

• stops
• fricatives
• affricates
• nasals
• glides
• liquids

by a complete closure somewhere in the vocal tract

• closure
• release
• transition
• reverse the steps for postvocalic stops

• corresponds to the complete closure of the vocal tract (slience)
• minimum radiated acoustic energy (silence for voiceless stops, voice bar for voiced stops, 50-150 ms*)

• pressure has been rising behind the obstruction
• rapid release produces a transient (20-30 ms)

• frication (air/noise)
• low freq. for /p/ (500-1500 Hz) (falling spectrum)
• high freq. for /t/ (above 4 kHz) (rising spectrum)
• mid-freq. for /k/ (1.5-4 kHz) (peaked spectrum)

a short amount of noise

• stop gap
• release of pressure
• transition

by voicing

• the interval between the release of the stop and the onset of vocal fold vibration
• for /b d g/ VOT from -20 to +20 ms with a mean of 10 ms
• for /p t k/ VOT from 25 to 80 ms with a mean of 45 ms

• VOT
• voice bar for intervocalic stops
• length of preceding vowel for final stops

because you are measuring from where voicing starts which is actually before the burst (it can be up to 20 ms behind where the burst is)- you make a sound before actually saying the consonant /b/ (called voice bar)

if they are followed by voiced consonants instead of voiceless

• articulatory movement from stop to vowel entails a formant movement
• as the resonating chamber of the vocal tract changes, the formant frequencies change
• formant transitions are important for perception
• formant transitions are approximately 50 ms in duration

you are changing your vocal tract

their vocal tracts are longer

usually rises

• for /p b/ F2 and F3 rise slightly
• for /t d/ F2 falls and F3 rises slightly
• for /k g/ F2 and F3 separate steeply and rapidly
• however, a given stop is associated with a variety of transitions (there is no fixed pattern for perception)

• narrow constriction in the vocal tract (not a complete closure like stops)- no moment of silence in spectrum
• when are flow rate is high, turbulence results
• turbulence is complex, unpredictable air flow
• turbulent airflow is perceived as turbulent noise
• fricatives have a relatively long duration

• sibilants (stridents)- greater noise energy (s, z, esh, yogh)
• nonsibilants (nonstridents)- f, v, theta, eth, h

airflow is neat and organized (predictable)

airflow is messy and unorganized, noise (fricatives)

• voicing- pulses (glottal closures) for /z yogh/, no pulses for /s esh/
• noise spectrum- alveolar sibilants have higher frequency energy range from 4 kHz to 12 kHz, palatal sibilants have energy down to 3 kHz, spectral irregularities aren't important in perception

formant transition locations depend on the articulation, but the transitions are not important perceptually for sibilants

• high
• /esh/ has more low frequency energy

• /f v theta eth h/
• less noise energy than sibilants
• voiced nonsibilants will have quasi-periodic pulses (some periodicity)
• noise spectra are fairly flat and diffuse

it is not known

formant transitions are primary acoustical cue (noise spectrum may play secondary role)

• described as a combination of stop and fricative
• /t-esh d-yogh/

• complete obstruction in the vocal tract
• intraoral pressure builds up
• release to generate fricative noise

• rise time- time it takes amplitude measure/envelope to go from 0 to maximum level
• duration of frication
• relative amplitude in third formant region
• stop gap- complete obstruction in vocal tract and it shows up as silence

• complete closure in vocal tract
• sound radiated through nasal cavities
• sometimes called nasal stops

sound of a nasal, acoustic waveform of nasal consonants

• nasal murmur
• associated strictly with nasal radiation of sound
• there are many spectral peaks, but most have low amplitude
• antiformants- loss of energy
• nasal formant (low frequency ~300 Hz, highest energy)

because higher formants have reduced energy

narrower than for vowels

• highly damped formants (broad bandwidths compared to other consonants)
• formant transitions in connected speech

because of mucous in nose

• also called approximants and semivowels
• /w j/

• gradual articulatory motion
• narrow, but not closed, vocal tract

• formants for /w/ F1 and F2 are both low
• formants for /j/ low F1 and high F2

F2 and F3

they have formants that we can see

• also included as semivowels
• /r l/

• rapid movements
• formant structure
• F3 is the main difference
• antiformants for /l/

closer together

separated

formant frequencies (ear looks for changes)

their disruption (either the complete closure or the constriction)

their antiformants (more than formants)