Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
88 Cards in this Set
- Front
- Back
|
morphology
|
how languages build words and indicate relationships between words
|
|
|
syntax
|
how words are organized into phrases, and phrases into sentences
|
|
|
semantics
|
how words and sentences convey meaning
|
|
|
pragmatics
|
how context affects meaning
|
|
|
phonetics
|
how to produce and perceive the sounds of the language
|
|
|
phonology
|
how the sounds of the language work together as a system
|
|
|
what do you know when you know a language?
|
- vast amount of systematic knowledge
- little conscious access to it - explicit vs. implicit knowledge |
|
|
explicit knowledge
|
- found in books, manuals, easy to access
- can be articulated with words and numbers - can be readily transmitted to others - things you know that you know |
|
|
implicit knowledge
|
- learned in an unstructured, formal way
- knowledge you are not necessarily conscious of - hard to articulate with formal language (though not impossible) - skills often captured with the term "know how" |
|
|
linguistic knowledge
|
- implicit
- automatic = you can produce word forms, sentences without giving it much thought - untaught = not explicitly taught to children - learned = need to learn a lot about the properties of your ambient language; acquired without instruction, through immersion |
|
|
levels of linguistic sound structure
|
- concrete...abstract
- continuous...discrete - signal...code - physical...mental (phonetics...phonology) |
|
|
what is the relation between the mental representation of speech sounds and their physical realization?
|
mental processes
|
|
|
phonetics (study of:)
|
- articulation & physiology of speech production
- acoustic characteristics of speech - processes by which speech is perceived - studied by phoneticians, linguists, psychologists, engineers, speech language pathologists |
|
|
phonotactics
|
knowledge about sound structure that allows you to legally combine sounds in a language
|
|
|
the speech chain
|
linguistic level - physiological level - acoustic level - physiological level - linguistic level
|
|
|
generation of speech sounds
|
- planned
- represented as a complex of abstract phonetic features - instantiated as neural commands to articulators - movements of articulators shape vocal tract - shape of vocal tract determines sound output |
|
|
production of all speech sounds is the result of...
|
modulation of the airflow from the lungs
|
|
|
3 systems of speech production
|
- supralaryngeal system/vocal tract (nasal cavity, oral cavity, pharynx)
- laryngeal system/larynx (and the glottis) - respiratory system/subglottal (trachea, bronchi, lungs) |
|
|
3 phases of speech production
|
- subglottal/respiratory = respiration
- laryngeal/larynx = phonation - supralaryngeal = articulation |
|
|
egressive speech sounds
|
- produced with exhaling
- most speech sounds are this |
|
|
inhalation in quiet breathing (process)
|
- medulla oblongata sends commands to resp. muscles
- diaphragm contracts (thoracic cavity expands vertically downward) - external intercostals, interchondral portions of internal intercostals, contract (thoracic cavity expands up and out) - lung volume increases b/c of pleural linkage - pressure decreases - air flows in through nose and mouth |
|
|
passive expiration in quite breathing
|
- relaxation of resp. muscles with air in the lungs:
- allows lungs and rib cage to recoil - resp. system collapses - pressure increases - air flows out - lungs return to resting volume |
|
|
active expiration (speech & singing)
|
- muscles counteract passive collapse of resp. system
- inspiratory muscles maintain lungs in expanded state - slow expiration early during exhalation phase - muscles force resp. system into compressed state - expiratory muscles compress thorax and abdomen - maintain expiration longer |
|
|
speech breathing & phrasing
|
- inspirations usually occur at major linguistic boundaries (phrases, sentences)
- long utterances require muscle control to maintain pressures throughout - utterance requirements affect both inspiratory and expiratory muscle use |
|
|
phonation
|
- airstream needed to set vocal folds in motion
- airflow manipulated for sound production - the vibration of vocal folds provides the voice source - creation of voice source converting air supply from lungs into audible vibrations for speech - larynx "voice box," trachea, vocal folds |
|
|
glottis
|
space in between vocal folds
|
|
|
vocal folds
|
- abducted (open) during normal breathing
- evolved to close off trachea when swallowing (epiglottis also involved) - evolved to trap air in the lungs for greater upper body strength - adducted (closed) for certain sounds - adducted with less resistance to airflow for voicing - partial abduction: whisper |
|
|
fundamental frequency (phonation)
|
- basic rate of vocal fold vibration
- physical property - measured in Hz - perceived as pitch |
|
|
fundamental frequency depends on...
|
- age, sex, language, paralinguistic reasons (mood, excitement..)
|
|
|
what controls changes in F0?
|
- vocal folds can be altered in three ways, changing vibrating frequency:
- thyroid cartilage can be tilted forward stretching the vocal folds - position of the arytenoid cartilages can be moved, changing the length and tension of the vocal folds - vocal folds themselves are part muscle and can be tensed |
|
|
articulation
|
- voice source created by phonation further manipulated by articulation and resonance
- = shapes of the vocal tract |
|
|
supralaryngeal tract includes
|
- lips, teeth, alveolar ridge, alveolpalatal region, hard palate, soft palate/velum, uvula
- tongue: tip, blade, body (front, center, back), root |
|
|
orbicularis oris
|
muscle involved in rounding the lips
|
|
|
phonetic transcription
|
- complete and structured list of the sounds of a language
- allows trained people to use it to transcribe the sounds of any language - listen to sounds "objectively" - minimize influence of native language/dialect - accurate transcription should allow any practiced reader to produce native sounding speech in any language/dialect |
|
|
International phonetic alphabet
|
- guiding principle: one sound = one symbol
- International Phonetic Association founded in 1886 in France, mostly by language teachers - different symbol for each distinctive sound - same symbol should be used for that sound in every language which uses it |
|
|
vowel
|
- "core" of a syllable
- produced with an unobstructed flow of air from the lungs - number of vowels in a language determined by minimal pairs |
|
|
parameters of vowel articulation
|
- position of tongue body: tongue height (high or low relative to the palate), tongue backness (front or back relative to lips & pharynx)
- advanced or retracted tongue body - lips (rounded, spread, neutral) - nasality (air passes/does not pass through nose as well as mouth) |
|
|
AE front vowels
|
i, ɪ, ɛ, æ
|
|
|
AE back vowels
|
u, ʊ, a
|
|
|
vowel space
|
- area in oral cavity within which the tongue can move without creating friction
- defined by position of two extreme vowels - i (front, high) & a (back, low) |
|
|
monophthongs
|
vowels that have the same quality throughout their production
|
|
|
diphthongs
|
vowels that change quality during their production
|
|
|
tense vowels
|
- i, e, o , u
- involve more extreme articulations - have longer durations - can occur in open syllables (ex. CV) - may be diphthongized |
|
|
lax vowels
|
- ɪ, ɛ, ʌ, ʊ
- are shorter in duration - occur only in closed syllables |
|
|
rhotacization
|
- vowels that are produced with an r-coloring quality
- can be achieved in two ways: with tongue tip raised (and curled) as in a retroflex consonant, with tongue tip down but tongue body bunched up |
|
|
nasalized vowels
|
- produced with soft palate lowered to allow part of airstream to go through nasal cavity
|
|
|
two visual representations for speech
|
- waveforms (time by amplitude/air pressure change)
- spectrograms (time by frequency (+amplitude)) |
|
|
sound
|
- variation in pressure over time
- set of movements, disturbances - vibration of air molecules that can cause the ear drum to vibrate, producing an auditory sensation - air pressure fluctuation |
|
|
simplest form of vibration
|
simple harmonic motion (simple movement - back & forth)
|
|
|
sine wave
|
- simplest kind of wave
- periodic wave that has a smooth and symmetrical curve connecting maximum and minimum pressure points - pure tone |
|
|
amplitude
|
- maximum amount of displacement of a particle on the medium from its rest position
- magnitude of the pressure variation in the wave - in a sound wave - diff. between peak pressure & average atmospheric pressure - usually measured in decibels (indicates intensity or power of sound wave) - perceived as loudness (greater amplitude = louder) |
|
|
cycle
|
complete round trip in a wave
|
|
|
period
|
- amount of time require to complete one full cycle
- usually measured in sec or msec - determined by physical properties of vibrating system (length of pendulum string, size, stiffness of tuning fork) - p = 1/F |
|
|
periodic
|
- when an event occurs repeatedly
- wave is periodic when the same pattern of air pressure variation is repeated exactly |
|
|
frequency
|
- number of times a wave repeats (cycles) per unit of time
- f = 1/T - perceived as pitch (higher frequency = higher pitch) |
|
|
infrasound
|
a sound with a frequency below the audible range of hearing
|
|
|
ultrasound
|
a sound with a frequency above the audible range of hearing
|
|
|
complex waves
|
- any wave that isn't a sine wave
- speech sounds are complex (made up of several frequencies) - periodic |
|
|
Fourier analysis
|
- all sounds can be analyzed by breaking them down in sinusoids which constitute the signal
|
|
|
Fourier synthesis
|
all sounds can be synthesized by adding sinusoids
|
|
|
to superimpose/add two waves...
|
add together the pressure/amplitude of each wave at each point in time (each sum is the pressure of the complex wave at that point)
|
|
|
representation of complex sounds
|
- complex periodic sounds can only have sinusoids that are multiples of its fundamental frequency
|
|
|
harmonics
|
- pure tone components of periodic sounds
|
|
|
fundamental frequency (acoustics)
|
- lowest frequency component
- represented in a spectrum (horizontal axis = frequency, vertical axis = amplitude) - first harmonic of complex wave (frequency at which the whole wave repeats) - frequency of simple wave produced by second mode is twice the F0, third mode is 3x F0... |
|
|
waveform vs. frequency spectrum
|
- waveform: time on x-axis, amp. on y-axis
- spectrum: frequency on x-axis, amp. on y-axis - waveform shows overall amp. of complex wave; spectrum shows amp. of each harmonic - waveform: individual harmonics not represented - spectrum: shows single time slice only, obtained by Fourier analysis |
|
|
aperiodic waves
|
- complex, but not harmonically related
- do not have repeating cycles - do not have harmonically related component waves |
|
|
resonator
|
- something set into forced vibration by another vibration
- acoustic resonator contains air - resonating frequency of a physical body is a freq. at which the body vibrates more efficiently than other freq. - bigger the object, lower its resonant freq. |
|
|
source-filter theory of speech production
|
- source in a sound is what sets air vibrating (in speech, vocal fold vibration & turbulence)
- voice source & its spectrum (airflow through vibrating larynx is sound source - opening & closing of vf "chops up" flow of air from lungs, creating complex wave) |
|
|
harmonics in the voice
|
- vocal fold vibration is rich in harmonics with F0 ranges from 100-300 Hz
- most speakers produce harmonics that have significant acoustic energy up to 4000-5000 - with lower F0, harmonics get closer together in the spectrum and cycles get further apart in the waveform |
|
|
filter - vocal tract
|
- supralaryngeal vocal tract can be characterized by a filter function, which specifies (for each freq.) the relative amount of energy that is passed through filter & out the mouth (reinforces some frequencies & dampens others)
- reinforced frequencies = formant frequencies |
|
|
formant frequencies
|
- peaks in the filter function of the vocal tract
- resonances of the vocal tract when it is in a particular configuration - act as frequency "windows" - allow specific freq. to pass through & bock transmission of others |
|
|
source-filter theory process
|
- source wave produced by vocal cord vibration - vocal tract filters source wave (allows some of component frequencies to pass through more strongly than others) - output spectrum has a harmonic energy distribution (but the amp. characteristics of harmonics have been shaped by formants)
|
|
|
source and filter characteristics can be changed...
|
independently
|
|
|
vowels & formants
|
- vowels produced with no significant constriction of oral tract
- articulations change resonances of vocal tract, which changes formant frequencies - formants give individual character to different vowels - frequency ranges (groups of harmonics) that emerge from mouth & nose with greatest relative amp. = resonant freq. of vocal tract |
|
|
spectrogram
|
- how the frequency spectrum changes over time
- time x frequency x amplitude |
|
|
types of speech noise visible in spectrograms
|
- phonation: vertical black striations at more or less regular time intervals; each striation reflects the noise made during a single glottal pulse
- fricative noise: messy stuff showing no regularities in the time domain; varies in darkness (amp.) from very black (high amp) to faint grey (low amp) - silence: white space - stop bursts: sharp line (transient) represents abrupt release of high-pressure air as articulators rapidly open at end of closure of an oral stop |
|
|
wide band spectrogram
|
- analyses over short time windows (averages over frequencies)
- better time resolution; more info in time domain - show formant structure but smearing harmonics |
|
|
narrow band spectrogram
|
- analyses over long time windows (more detail visible with respect to frequency dimension)
- bad time resolution - reveal harmonic structure - hard to find formants |
|
|
F1
|
- inversely correlated with tongue height
- high vowels: low F1 (large pharyngeal cavity) - low vowels: high F1 (small pharyngeal cavity) |
|
|
F2
|
- correlated with tongue backness
- front vowels: high F2 (small oral cavity) - back vowels: low F2 (large oral cavity) - lip rounding lowers F2 |
|
|
things that change formant frequencies
|
- overall length of vocal tract (longer VT - larger resonants - lower frequencies)
- location of constriction (narrowing along length of vocal tract; back tongue constriction lowers F2) - degree of constriction (tighter constriction lower F1) |
|
|
sources of vowel variation
|
- information conveyed by vowels (vowel quality - vowel ID)
- across-talker variation (language, dialect, gender, historical change) - within-talker variation (speaking style, social group, contextual variation) |
|
|
vowels in different languages
|
- most common vowel systems include just 5-7 vowels (Spanish, Greek, Hawaiian)
- smallest known vowel inventory has just 3 vowels (Arabic) - English has 10+ - German: large vowel inventory |
|
|
what features can "add" vowels to a a vowel space?
|
- front/back (all languages have this)
- height (all languages have high vs. low; adding mid gets you from 3 to 5 vowel system) |
|
|
[a]
|
- most common low vowel in world's languages
- if a language has just one low vowel it is likely to be [a] |
|
|
[e]
|
- for most English speakers, it occurs in a diphthong
- common in other languages as simple monophthong |
|
|
lip rouding
|
- in many languages isn't distinctive (ex. English)
- it is often predictable from vowel backness (back vowels are rounded) - rounding is distinctive in a language like French - less common, back vowels can be unrounded (Vietnamese) |
|
|
why are front vowels almost always unrounded, while back vowels are almost always rounded?
|
- features front & unrounded both conspire to increase frequency of F2
- highest F2 found in vowels is in /i/ - conversely, back & round conspire to lower F2, so lowest F2 found in /u/ - lip rounding involves lip protrusion, which lengthens vocal tract, which lowers all formants |
