451

Three formants are required for perception of /r/ & /l/

F3 distinguishes /r/ from /l/ - F3 is the lowest for /r/ being between 1200-1800 Hz, while F3 for /l/ is between 2400-3000 Hz

In most vowel contexts, F3 for /l/ does not vary in frequency

O’Connor et al – synthesized perceptually acceptable /w/’s and /j/’s with only 2 Formants

F1 for both sounds is similar – very low

F2 distinguished these sounds as F2 for /j/ is much higher (2300 Hz) than for /w/ (900 Hz).

Important acoustic characteristics are the changing formant frequencies or transitions

Transitions occur when a vowel precedes or follows and reflects changes in VT shape

Have more rapid transitions than do diphthongs

The lowest F3 of all sound in English

• Formant frequencies:
• F1 320 Hz
• F2 1200-1800 Hz
• F3 1500 Hz
• F2 and F3 are narrowly separated

Well defined formant structure

Has higher F1 than glides

• Formant frequencies
• F1 360 Hz
• F2 1305 Hz
• F3 2400-3000 HzHas antiresonances

Syllable initial /l/ -articulated w/ dorsum of tongue low in mouth /liv/

Syllable final /l/ - articulated w/ dorsum raised toward velum /kul/ - called a dark /l/ & assoc’d w/ high back vowels

Syllable final /r/ - dorsum elevated towards velum - loses consonant quality & colors following vowel, producing a low F3 in the following vowel

• Primary articulatory characteristic:
• Tongue tip raised to alveolar ridge when syllable initial

Sound is directed through distinctively shaped oral cavity, one that can be held indefinitely for sustained production of sound

For /l/ - tip rests lightly on alveolar ridge & divides airflow into 2 streams over sides of tongue (lateral)

For /r/ the tongue is grooved & doesn’t touch alveolar ridge. Some of airflow emerges centrally but most emitted laterally.

• Formant frequencies similar to /i/:
• F1 300 Hz
• F2 2300 Hz
• F3 3000 Hz
• Very low F1

• Formant frequencies similar to /u/:
• F1 300 Hz
• F2 900 Hz
• F3 2200 HzVery low F1

Primary acoustic feature:

• 1) Slowly changing formant patterns
• 2) Transition durations of 75-150 ms
• 3) Formants glide up or down to/from the preceding or following vowels
• 4) A very low F1 frequency

• 2) Position is close to that for the high front vowel /i/
• 3) Is palatal

• /w/ has 2 regions of narrowing
• 1. lips
• 2. Between the dorsum of tongue and palate
• 3. Labiovelar
• 4. Lip & tongue movements are in close coordination, beginning and ending together
• 5. Similar VT configuration as for high back vowel /u/

Are both referred to as ‘semi-vowels’ or ‘resonant’ consonants.

Called semi-vowels because formant structure is similar to vowels

Both /r/ and /l/ can be sustained and ‘resonated’

BUT they’re not vowels because they occur on syllable periphery and not as syllable nuclei unless they functioning as a syllabic consonant

• Primary articulatory characteristic: gliding motion of the articulators
• 1) VT configuration somewhat narrower than for vowels but less constricted than for stops or fricatives
• 2) Characterized by a gliding motion of the articulators from a partly constricted state to a more open state
• 3) Gliding movements are slower than the opening and closing movements for stops.

Nasal murmurs are between 200-300 Hz

But, they vary somewhat w/ place of articulation

They are lowest for /m/, then /n/, then /ŋ/

Occur due to the occlusion of the oral cavity and its function as a cul de sac resonator

Are frequency regions where the amplitudes of components of the glottal source are severely attenuated

If they’re the same frequencies or close, resonances and antiresonances cancel each other out

• Nasals are weak sounds due to
• 1) Antiresonances

2) The effects of a longer vocal tract

3) Sound absorption that occurs due to the passage of sound through the nasal cavities

Open VP port allows for oral and nasal cavity coupling

Coupling results in a ‘cul de sac’ resonator, the oral cavity

A cul de sac resonator: 2 cavities are coupled and one has no external aperture (which?)

Within the cul de sac resonator (oral cavity) antiresonances occur.

Absence of energy at a given frequency can indicate the volume (size) of the oral cavity

Opening the VP port and adding the nasal cavities to the vocal tract w/ occluded oral cavity results in:

1. A longer/larger VT which = lower resonant frequencies

2. Oral cavity functions as a ‘cul de sac’ resonator, & is responsible for the ‘nasal murmur’

• 3. Nasal murmurs are between 200-300 Hz
• are lowest for /m/, then /n/, then /ŋ/

Is capable of modifying the vocal tract by either

• 1) Closing for vowels & oral consonants to incr oral cavity pressure
• 2) Opening for nasal sounds w/ oral cavity is occluded - allows for nasal cavity resonance

Degree of closure varies: >’est for oral consonants, such as stops, affricates, & fricatives, then high vowels, then low vowels; least for nasals

• nVelum w/in 2 mm of the pharyngeal wall - no nasality perceived
• nVelum w/in 5 mm of the pharyngeal wall, nasality perceived

Velar height is related to incr size of the supraglottal cavity. High velum incr supraglottal cavity. Facilitates VF vibration during voiced stop production

Incr supraglottal area, decr supraglottal pressure, and this allows VFs to vibrate for voiced stop production

• Primary articulatory features:
• Completely occluded oral cavity
• Open velopharyngeal port
• Sound is radiated through the nasal cavities

• Primary acoustic features
• Nasal murmur
• Antiresonances

• Are produced by constricting the VT
• a. A narrow constriction in the VT requires the close approximation of 2 articulators

The primary articulatory feature is this narrow constriction and the turbulent airflow that it produces


Air stream is sent through narrow constriction in VT

Airflow must be strong enough & constriction narrow enough to create turbulent air flow

Turbulent air flow creates aperiodic sound that may or may not be accompanied by a periodic sound, VF vibration.

The aperiodic sound is resonated through the VT

The most effective resonator is that which is anterior to the point of constriction

• Primary Acoustic Feature:
• 1. Continuous Frication Noise
• 2. Little or no formant structure

1. Sibilants or stridents /s, z, ʒ, S / Higher energy, that is, greater intensity, due to a larger anterior cavity. The anterior cavity is the area in front of the point of constriction.

2. Non-sibilants or non-stridents /f,v,q,ð,h/ Lower energy, less intensity, due to smaller anterior cavity

• Both have continuous noise as primary acoustic characteristic related to manner
• But have different spectrographic characteristics related to place of articulation

PHASE 1 - Closing Phase

• 2 simultaneous occlusions:
• 1. Oral cavity occlusion via lip closure, tongue-alveolar ridge articulation, or tongue-palate articulation 2. Velopharyngeal port occlusion

PHASE 2 – Hold or closure phase

• Occlusion is maintained at the point of VT obstruction, no air flows through VT
• 1. Intraoral air pressure builds
• 2. Must have intraoral pressure build-up for stop production

• PHASE 3 – Release phase
• * One of the 2 occlusions, usually the oral occlusion, is broken and air is released
• * Release by lowering velum – is used when a stop is followed by a ‘homorganic’ nasal or ‘homorganic’ syllabic nasal. Homorganic means stop and nasal phoneme are produced in the same place. Example: ‘hidden’

1. ‘Stop gap’ or ‘silent gap’

• Period of silence during the hold phase
• No air flow
• Devoid of formant structure or noise
• Not always silent :
• For /b, d, g/ - there is sometimes a low intensity Fo

2. Noise Burst

• Occurs at moment of stop release
• Vertical spike on spectrogram
• Duration: 10-35 ms
• Covers broad range of frequencies
• The most intense frequencies in the burst are related to place of articulation

3. F1 transition

• The change in F1 (frequency of first formant) that occurs as the VT changes shape after the release of initial stops and before the occlusion of final stops
• F1 incr in frequency rapidly after release for initial stops, and falls rapidly before occlusion for final stops
• nHow does this relate to mouth opening?

1. Noise bursts :burst of transient noise that occurs at stop release

nBilabials 600Hz, Alveolars 3000Hz, Velars vary – are related to F2 of following vowel

2. Direction of F2 transition – second formant frequency transition

• Direction of the F2 transition (change in F2 frequency) from stop to following the vowel or from vowel to following stop.
• Each stop has a typical F2 locus: /p, b/ 600 Hz, /t, d/ 1800 Hz, /g, k/ depends on the vowel
• Rapid movements of articulators cause a sudden change in formant frequencies of VT
• F2 reflects the movement of the tongue or lips from or to a place of stop occlusion