question

Source

answer

1. VF Vibration 2. F0 and harmonics 3. Pitch

question

Filter

answer

1. Vocal Tract 2. Formants 3. vowels/some other voiced sounds

question

Resonance

answer

Vibratory response to an applied source

question

Resonator

answer

- Something set into forced vibration by another vibration - Acoustic resonator - air filled container - Act as filters, amplifying some sounds and weakening others

question

Natural Resonant Frequency

answer

- Determined by physical properties of the object such as mass, shape, and tension - frequency at which an object oscillates with the greatest amplitude

question

Constructive Interference

answer

- frequency component matches the container's resonant frequencies - amplification occurs

question

Destructive Interference

answer

- frequencies other than NRF - attenuation occurs

question

High Pass Filter

answer

- allows high frequencies to pass (low-pass does the opposite)

question

Bandpass Filter

answer

-Permits middle range of frequencies to pass while blocking high and low

question

Bandwith

answer

- range of frequencies that a given resonator will transmit - small bandwidth = narrowly tuned - large bandwidth = broadly tuned

question

Vocal Tract Filter Properties

answer

- broadly-tuned - bandpass

question

Fundamental Frequency

answer

- rate of VF vibration

question

Harmonics

answer

- Whole number multiples of F0 - unlimited, but amplitude falls off at higher frequencies

question

Variable Properties of Vocal Tract

answer

- Space in oral cavity can be wide or narrow - Coupling of nasal cavity - constrictions divide the VT into larger and smaller cavities which resonate with their own frequencies

question

Formants

answer

- resonant frequencies of the VT - depend on the shape of the VT (filter) - independent of F0 (source)

question

LPC Spectrum

answer

- shows the spectrum of the voice after filtering through the VT - used in biofeedback therapy for sonorant targets; clinician can make a template to act as a target - CSL Sona-Match - changes in real time as speaker produces different sounds - important to use template from someone with similar size VT to client

question

Populations that benefit from vowel spectra in therapy

answer

- patients with hearing loss - ESL speakers - children with CAS or functional articulation disorder

question

Narrow Band Filter

answer

- Good fx resolution; less good time resolution (smearing) - show individual harmonics of voiced sounds, can't see formants - useful for tracking F0 - Filter set to 45 Hz

question

Wide Band Filter

answer

- Less detailed Fx resolution; good time resolution - resonances of the VT are well defined; individual harmonics are lost - useful for tracking formants - filter set ~300 Hz

question

Point Vowels

answer

- /i/ /a/ /u/

question

F1

answer

- Inversely correlated with tongue height - High vowels = low F1 - Low vowels = high F1 (tongue root forced back into pharyngeal cavity creating smaller volume - determined by volume of pharyngeal cavity

question

F2

answer

- Inversely related to tongue backness - Front vowels have a short front cavity = high F2 - Back Vowels have long front cavity = low F2

question

Source-Filter Theory

answer

1. VF Vibration (source) creates a complex wave 2. Supralaryngeal VT acts as a filter, resonating at certain frequencies

question

Complex Wave

answer

- F0 and its harmonics

question

Resonance and Volume

answer

- large container = low frequency - small container = high frequency - change size of oral and pharyngeal cavities to change vowel quality

question

High Front Vowels

answer

- low F1, high F2 - F1 & F2 very far apart

question

Low Back Vowels

answer

- High F1, Low F2 - F1 and F2 close together

question

Round Vowels

answer

- most back vowels in english; no front - rounded lips make oral cavity even bigger for back vowels, making F2 even lower - makes it even easier to hear the difference between front and back

question

Offglide

answer

- articulatory end point of the diphthong - /I/ or /ʊ/

question

Glides/Semi vowels

answer

- look like their corresponding vowels on a spectrogram - palatal glide [j] looks like [i] - labio-velar glide [w] looks like [u]

question

3 stages of Consonant Production

answer

1. Shutting movement/ closing phase 2. Closure 3. Release of Closure

question

Formants and Consonants

answer

- F1 always low next to stops bc stops have the highest tongue position - F2 direction of formant movement depends on the consonant place of articulation and the vowel quality

question

Causes of Formant Variation

answer

- Anatomical: gender, age - larger VT = lower formants - Dialectical

question

Articulatory Undershoot

answer

- Articulator movements fall short of normal target location - vowel space area is reduced - significant negative impact on intelligibility - Result of motor speech disorder - (measuring vowel formants)

question

Parkinson's Disease

answer

- articulatory undershoot common - speech often perceived as abnormally rapid even when measurements reveal rate is within normal limits - in typical speakers vowel space is compressed in rapid speech, so listener's associate undershoot with rapid rate - use instrumental measures to determine if intelligibility problem is due to rapid rate or undershoot

question

Obstruent Consonants

answer

- sound source: created by turbulence as air passes through narrow constriction in oral cavity (supraglottic/supralaryngeal) - supraglottic sound source is aperiodic

question

Fricatives

answer

- characterized bt aperiodic noise distributed over a range of frequencies - higher frequencies than for other sounds (set praat to 8000 Hz)

question

Non-strident Fricatives

answer

- low intensity energy that is evenly distributed across frequencies - note f sometimes strident but not sibilant

question

Strident Fricatives

answer

- more intense energy that is concentrated at higher frequencies

question

Fricative Voicing

answer

- Spectrogram: voicing bar - Waveform: aperiodic noise superimposed on periodic signal of VF vibration; periodic component tends to have greater intensity

question

Fricatives: Place

answer

- length of oral cavity determines frequency - longer cavity has lower resonant frequency

question

Acoustic components of stop

answer

- silent gap - release bust - VOT - Formant transition

question

Release Burst

answer

- explosion of aperiodic sound following silent gap - turbulence noise created by rush of pressurized air - looks like fricative noise on spectrogram but much shorter duration - may be absent from word final (unreleased) stops

question

Velar Pinch

answer

- 2nd and 3rd formants approach each other for velars

question

Affricates

answer

- homorganic stop-fricative sequence - rise time, rate of amplitude increase of frication noise, distinguishes affricates (short) from fricatives (long)

question

Biofeedback: Fricatives

answer

- use real time feedback to target /s/ /sh/ contrast

question

VOT to determine Source of Error (s deleted from stop clusters)

answer

- Phonetic/planning issue (apraxia) : trouble physically producing cluste; motor learning helps - Phonological Issue (aphasia): trouble at level of representation; motor learning does not help

question

Sonorant Sound Source

answer

- VF vibration - greater obstruction blocking the escape of sound results in lower intensity than vowels

question

Nasals

answer

- velum lowered, velopharyngeal port open - periodic voicing - low frequency 1st formant - lower amplitude than adjacent vowels - look like low intensity vowels

question

Nasals: Intensity

answer

- loss of intensity bc soft, mucus and cilia lined nasal cavities absorb sound - small apertures (nostrils) do not transmit sound effectively - easily confused due to their low intensity

question

Nasals: Frequency

answer

- low frequencies more intense - coupled cavities form a large resonator -nasal formant = low F1

question

Anti-formants

answer

- sound resonating in closed off oral cavity gets trapped & cannot contribute to sound of phoneme - creates area of unusually low intensity (blank space) on spectrogram

question

Glides

answer

- like short vowels that pattern as consonants (occur at edges of syllabi not nuclei) - lack a clear steady state interval due to short duration (~ 75 ms) - may be difficult to distinguish from vowel. Look for short duration and lower intensity - j like high front vowel - w like u

question

Liquids

answer

- /l/ - sides of tongue down, air flows laterally - acoustics evident in F3 (low for r) - resemble vowels with lowered intensity - low F2

question

Rhotic Vowels

answer

- made with combination of labial, post-alveolar, and pharyngeal constriction - result is very low F3

question

Anticipatory Coarticulation

answer

- regressive - one segment is changed bc of a segment that occurs later

question

Perseveratory Coarticulation

answer

- progressive - some feature/gesture is maintained after the target segment - example: plural voicing assimilation

question

Hypernasality

answer

- too nasal - found in patients with cleft palate or velopharyngeal weakness - appears as a continuous band of low frequency energy on the spectrogram - affects entire utterance not just nasal sounds - can use acoustic signal to help client understand/eliminate hyper nasal voice quality

question

VOT in children

answer

- control develops late in development, around age 6 - voicing neutralization errors common

question

Covert Contrast

answer

- child produces reliable, measurable distinction between 2 sound - but does not realize contrast in same way as adult speakers; adults do not hear the contrast as child produces it - likely to resolve spontaneously; children with this reach criterion in fewer sessions - may be more efficient to focus on other errors

question

Perception (interpretation of sounds) based on:

answer

- linguistic background/phonology - surrounding speech sounds (context/coarticulation) - situational/semantic context - cues from other sensory modes

question

Segmentation Problem

answer

- we hear speech as discrete units but this is not the case

question

Invariance Problem

answer

- one to many problem: one acoustic signal can evoke different percepts (based on formants of carrier phrase) - many to one problem: different signals can evoke same precept - across speakers: signal effected by age, gender, size, dialect, emotional state, rate of speech, register

question

Categorical Perception

answer

- humans systematically perceive acoustic signals as belonging to one category or another - sharp crossover from one phoneme to another - does not become language specific until ~ 9 months

question

Native Language Category Boundaries

answer

- English p/b - crossover at +25 ms - Spanish p/b - crossover at 0 ms ("pure" voicing) - Thai ph/p/b: 2 crossover locations instead of 1

question

Infant Directed Speech

answer

- slower rate, higher f0, greater speech variation, longer pauses, repetitive intonation, simplified sentence structure - probably helps in perceptual learning - Expanded vowel space - exaggerated vowels are better separated in acoustic space which may help infants identify categories/boundaries

question

Social Gating Hypothesis

answer

- social interaction creates optimal environment for perceptual learning - may explain nature and cause of language impairment in children with autism, as these children prefer nonsuch stimuli to IDS - at 2 years show different EEG response to speech stimuli

Speech Science Final – Flashcards

Unlock all answers in this set

Haven't found what you were looking for?

Search for samples, answers to your questions and flashcards