Phonetics Final PDF

CONSONANTS - Phonemic Features What are consonants - Phonemes produced with noticeable constriction or narrowing of the vocal tract (more constricted than vowels) - If the tongue is involved, it is possible to identify a very specific segment of the tongue creating the constriction...

CONSONANTS - Phonemic Features What are consonants - Phonemes produced with noticeable constriction or narrowing of the vocal tract (more constricted than vowels) - If the tongue is involved, it is possible to identify a very specific segment of the tongue creating the constriction - Most consonants involve the tongue - Rarely serve as the nucleus of a syllable - Only l, m, and n can be syllabic - Velopharyngeal port is usually closed - Except for nasals - m, n, “ng” Basic phonemic features of consonants - Manner of articulation - the way in which the consonant is produced - Place of articulation - the location in the mouth where the major constriction occurs - Voicing - whether or not the vocal folds are vibrating during phonation Manner of articulation - How a consonant is produced - Only 6 (7) used in english - Stop - complete closure, stops airflow - Nasal - allows energy to come out nose - Fricative - air flows around or through tongue - Affricate - combination between a stop and a fricative - Liquid - Lateral - “l” sound - sound flows out sides of tongue - Rhotic - “r” sound - pulls the tip of the tongue back - Glide - tongue moves from one position to another quickly - always followed by vowel Place of articulation - Location in the mouth where the major constriction occurs - Many possible places in the vocal tract - Only 7 (sometimes 8) used in English - Bilabial - using both lips - Labiodental - teeth and lips - Dental - tongue against or through teeth - Alveolar - tongue against alveolar ridge - Post alveolar sometimes used by linguists, not SLPs - Palatal - tongue approaching hard palate - Palatal-velar (aka velar) - constriction somewhere between hard and soft palate - Glottal - sounds involving constricting or closing glottis Voicing - Whether or not the vocal folds are vibrating during production - Voiced - vocal folds vibrating - Voiceless = no vocal fold vibration Using phonemic features to describe consonants - All consonants can be defined by manner, place, voicing (in that order) - Examples - /p/ - stop, bilabial, voiceless - /s/ - fricative, alveolar, voiceless - /j/ - glide, palatal, voiced - Theoretically in English there are 98 possible combinations of manner, place, and voicing - However, not all possible combinations are used - Every language uses a subset - The subset is called the inventory - Changes in each feature result in changes in meaning - which is why they are “phonemic” features MANNER OF ARTICULATION Stop consonants - Phonemic - /p/, /b/, /t/, /d/, /k/, /g/ - Non-phonemic - /ʔ/ - termed a glottal stop - Can be:  - Used to separate a word that ends in a vowel from one that starts with a vowel - A compensatory articulation for a poorly functioning velopharyngeal port - Would be used in place of most or all stop consonants - Referral to MD for evaluation would be indicated - An allophonic variation when a voiceless stop consonant precedes an unstressed syllable - And that unstressed syllable has a nasal after the syllable nucleus - Ex: “important” /ɪmpɔrʔənt/ - How are stop consonants produced - Stop consonants are dynamic and involve 2-3 phases in their production 1. Stop or closure phase a. The tongue, lips, or vocal folds create a total constriction in the vocal tract b. Velum raises, closing the VP port, and prevents air from entering the nasal passages c. Air pressure builds up behind the constriction 2. Release or burst phase a. The stop closure is released causing the built up air pressure to be released resulting in a very short burst of noise 3. Aspiration - voiceless aspirated stops only a. Clusters and voiced consonants will not have aspiration b. During phase 1 (closure) the vocal folds are abducted c. After phase 2 (release) the vocal folds begin to adduct for production of the following vowel i. As the vocal folds adduct a turbulent noise is produced called aspiration - Commonly used names for stop consonants - Stops - refers to the phase 1 closure - Plosives - refers to the phase 2 burst - Stop-plosives - refers to both the phase 1 closure and phase 2 release burst - A burst does not always occur - This is especially true when stop consonants occur at the ends of the words - In this case stop is the more accurate term - Stop consonants are among the most rapid movements in speech - Both the closure and the release phases are rapid A special case of stop consonant: the flap - What is it? - When the tip of the tongue makes a light contact with the alveolar ridge - Does not build up the pressure to crease a phase 2 burst - Represented by IPA symbol /ɾ/ - Also called a “tap” or a one-tap trill - When does a flap occur  - Occurs only with stop consonants /t/ and /d/ when between two vowels (intervocalic) - Almost always between a stressed followed by an unstressed syllable - Examples: - Letter [ǀɛɾɚ] - Middle [mɪɾəl] - Padding [pæɾɪŋ] Fricative consonants - All are phonemic - /f/, /v/ /θ/, /ð/, /s/, /z/, /ʃ/, /ʒ/, /h/ - How are fricative consonants produced? - By making a very narrow constriction in the vocal tract (but not complete closure) - The narrow constriction makes air spiral around in a turbulent eddy - The turbulent eddy creates turbulent noise - A fricative consonant is turbulent noise - How is a fricative consonant different from the burst of a stop consonant - A burst is a noise created by the sudden release of a constriction during a stop consonant and is very short - A fricative consonant is turbulent noise created by air passing through a constriction in the vocal tract and lasts much longer - How is a fricative consonant different from the aspiration phase of a stop consonant - The aspiration phase of a stop consonant is turbulent noise created by vocal folds as they adduct for the following vowel - A fricative consonant is turbulent noise created by air passing through a constriction in the vocal tract - Sibilants - Most intense fricative - /s/, /z/m /ʃ/, /ʒ/ - Non sibilants - Least intense fricatives (non-sibilants) - /f/, /v/, /θ/, /ð/, /h/ - /ʒ/ - Least frequent of English sounds - Never occurs at the beginning of English words Nasal Consonants - Phonemic - /m/, /n/, /ŋ/ - How are nasal consonants produced? - Airflow through the mouth is completely blocked at either the lips, the alveolar ridge, or the velum - Velopharyngeal port is open and sound energy is directed out of the nose - All nasal consonants are voiced (produced with vocal fold vibration) - /m/ and /n/ can be nucleus of a syllable - they can be syllabic  - Examples: spasm, button Liquid consonants - Phonemic - /r/, /l/ - How are liquid consonants pronounced - Vowel like consonants in which the vocal tract is constricted only slightly more than for vowels - All are voiced - Two subtypes in English - Lateral - Rhotic - How is lateral /l/ consonant pronounced? - Tongue creates a closure along the roof of the mouth at the midline - Tongue tip touches at the alveolar ridge - Sides of tongue do not contact roof of mouth - Sound energy flows around the sides of the tongue and out of the mouth - Rhotic “r” - Note on the differences between traditional IPA and the modified IPA used by SLPs when representing the rhotic consonant - Traditional IPA - American rhotic “r” = /ɹ/ - Trilled romance language “r” = /r/ - Modified IPA used by SLPs - American rhotic “r” = /r/ - Trilled romance language “r” = /r̗/ - How is the rhotic /r/ pronounced? - Likely many different ways to produce this sound - Two descriptions are the most common and are the basis for most clinical intervention - Retroflexed - tongue tip is curled up backwards, pointing toward the palate - Bunched - tongue tip points down and the body of the tongue is crowded back and up toward the palate - In all versions there is also narrowing of the pharynx which may contribute significantly to the sound quality - Important things to remember - R-like sounds can function as both a consonant /r/ or a vowel /ɚ/ and /ɝ/ - How we transcribe it depends on the role it serves in the word being produced - If the r-like sounds rhymes with “her” it is a vowel - If not is it a consonant Glide consonants - Phonemic - /j/, /w/  - Non-phonemic - /ʍ/ - aspirated “w” - Sometimes called semi-vowels - How is a glide consonant produced? - Involved gliding movement of the tongue from a partial constriction of the vocal tract to a more open vocal tract for the following vowel - This differs from a diphthong that is a movement from one vowel position to another - By definition, must be followed by a vowel - Never occur at the end of a word - Ex: hollow, shadow, holiday - all end in vowel sounds Affricate consonants - Phonemic - /ʧ/, /ʤ/ - Must be written with symbols touching to show that they are ONE phoneme - How is an affricate consonant produced - Combination of a stop closure and a fricative segment - Usually produce at the same or nearby place of articulation - Fricative segment is shorter than a full fricative consonant - In English fricative portion only produced at ‘palatal” place - In IPA these are described as “palato-alveolar” Additional manner terms - Two broad categories of consonants are sometimes distinguished - Obstruents - Includes all stops, fricatives, and affricates - Consonants produced with either complete or very narrow constriction of the vocal tract - Sonorants - Includes all liquids, glides, and nasals - Consonants produced with relatively less constriction of the vocal tract than obstruents but more than vowels - One term for liquids and glides together - Approximants - Homotypic - When sounds are produced with the same manner of articulation, they are said to be homotypic - Example /f, v, s, z, ʃ, ʒ, h/ are all fricatives, so they are all homotypic PLACE OF ARTICULATION Bilabials - 5 English sounds produced by the partial or complete coming together of both lips - /p/, /b/, /m/, /w/, /ʍ/ - Bilabial consonants include: /p/, /b/, /m/  - Bilabial stops: voiceless /p/ and voiced /b/ - Lip closure = phase 1 of stop consonant production - Velopharyngeal port is closed - Voiced bilabial nasal /m/ - Lips are closed throughout production - Velopharyngeal port is open - Sounds energy flows out of the nose - Labio-velar glide: voiced /w/ and voiceless /ʍ/ - Sometimes called bilabial because both involve lip rounding - Also classed as velar because the tongue creates and initial constriction at the soft palate, then moves toward the following vowel Labiodentals - Voiceless /f/ and voiced /v/ - Lower lip raised to approximate, the upper incisor teeth - Airflow is directed across the upper surface of the lower lip - Narrow constriction generated turbulent airflow - Sound energy (noise) created by /f/ and /v/ is much weaker than for /s/ or /z/ Dentals - Voiceless /θ/ (theta) and voiced /ð/ (eth) - The tongue tip contacts the back of the upper incisor teeth - May also be formed by protruding the tongue tip slightly between the teeth - These sounds are sometimes called interdentals - Airflow is directed between the upper surface of the tongue and the teeth - Narrow constriction creates turbulent airflow - Sound energy created by /θ/ and /ð/ as weak as for /f/ and /v/ - Usually the last sounds to be learned by most children - they are hard to articulate Alveolars - Includes 6 English sounds - /t/, /d/, /s/, /z/, /l/, /n/ - Voiceless /t/ and voiced /d/ - Tongue tip completely closes off the airstream at the alveolar ridge - Closure phase for /t/ tends to be longer in duration than /d/ - Exact tongue position for /t/ and /d/ tends to vary with context - Example: /d/ produced on teeth when just before a dental fricative as in “width” /wɪdθ/ - Example: /d/ produced with blade of tongue when just before a palatal as in “dues” /djuz/ - Voiceless /s/ and voiced /z/ - Tongue tip places near alveolar ridge with sides of tongue against upper teeth - Noise is created as air flows down narrow groove formed in center of the tongue  - Some speakers produce these with tongue tip down contacting the lower from teeth - Air flows through a narrow groove in center of the tongue blade - Voiced liquid /l/ - Tongue tip against alveolar ridge - Air flows around sides of tongue - Can serve as syllable nucleus (can be syllabic), most likely in fast speech - Ex: bottle, sizzle, navel - Possible light and dark allophones - Generally prevocalic = light - touches alveolar ridge - Postvocalic + syllabic = dark - only approximates alveolar ridge - Voiced alveolar nasal /n/ - Closure with tongue tip at alveolar ridge like /t/ and /d/ - Velopharyngeal port open - Sound energy flows out the nose - Timing of the opening/closing of port determines its effects on adjoining sounds (coarticulation) Palatals - Includes 6 English sounds - /ʃ/, /ʒ/, /ʧ/ˌ /ʤ/, /r/, /j/ - Palatal fricatives voiceless /ʃ/ and voiced /ʒ/ - Tongue blade raised close to the palate - Narrow constriction created turbulent airflow as air passes between the tongue blade and the palate - Tend to be produced with lip rounding - /ʃ/ - symbol sometimes called “esh” - Palatal affricates voiceless /ʧ/ and voiced /ʤ/ - Only true affricates in English - Place of articulation is based on the second sound, the palatal fricative - Voiced palatal rhotic /r/ - Many different articulations likely - A few common allophones - /r̟/ = fricative version of /r/ - Seen after /t/ or /d/ - /r̗/ = trilled version of /r/ - Most common after /θ/ - /ɾ/ = “flapped” /r/ - in British English - “Very” sounds like “veddy” - Voiced palatal glide /j/ - Involved movement from narrow constriction near the palate to the position for the following vowel - It must be followed by a vowel and never occurs in word final Velars - Includes 5 English sounds  - /k/, /g/, /ŋ/, /w/, /ʍ/ - Velar stops voiceless /k/ and voiced /g/ - Tongue dorsum closes off the vocal tract at the velum - Velopharyngeal port closed - Voiced velar nasal /ŋ/ - Tongue dorsum closes off the vocal tract at the velum - Velopharyngeal port open - Occurs in words where “n” precedes either “g” or “k” - In English /ŋ/ never occurs at the beginning of a word - Before the letter “g” - “G” may not be pronounced - /ŋ/ has a common allograph of “ng” where the “g” is typically not pronounced - Ex: “sing” = /sɪŋ/ - “G” may be pronounced - Ex: “finger” = /fɪŋgɚ/ - Before the letter “k” - “K” is pronounced after /ŋ/ - Labio-velar glide voiced /w/ and voiceless /ʍ/ - Classified as velars because the tongue body of the tongue approximated the velum - Similar to the position of /u/ but higher - It is the movement away from this position towards the following vowel that leads to perception of the glide Glottals - Includes 2 English sounds - /h/, /ʔ/ - Voiceless glottal fricative /h/ - Vocal folds create a narrow constriction - Creates turbulent airflow - Never occurs in word-final position - Because glides also never occur in word-final position, some linguists refer to /h/ as a glide - not for SLPs - Voiceless glottal stop /ʔ/ - As in “oh-oh” - Vocal folds close and open very rapidly - Occurs most frequently in connected speech between two vowels at word boundaries Additional IPA places not typically used in English - Post alveolar - Uvular - Pharyngeal IPA also called “retroflex” a place, we call it a manner Homorganic - when sounds occur at the same place of articulation VOICING 2 types of voicing - Phonetic voicing - Actual vibration of the vocal folds during some part of the consonant - Phonological voicing - Intention to produce a voiced phoneme or perception of a voiced phoneme whether or not actual vibration of the vocal folds has occurred Phonetic voicing - voices/voiceless contrast - Stop consonants - All stop consonants have a voiced and voiceless version - Fricative consonants - All fricatives but one have a voiced and voiceless version - Exception is the voiceless aspirate /h/ - Affricate consonants - All affricates have a voiced and voiceless versions - All obstruents have voiced and voiceless versions except for the fricative /h/ - No voice/voiceless contrast - Glides - all voiced - Liquids - all voiced - Nasals - all voiced Phonological voicing - Intention by speaker to produce a voiced or voiceless phoneme, or perception by listener of voicing whether or not physical vibration of the vocal folds occurs during the phoneme - Voicing or lack of voicing is phonemic in English - It changes the meaning of a morpheme - Ex: the speaker intends and the listener perceived the word “dad” even though the vocal folds did not vibrate during the final /d/ - How is phonological possible with no vibration of the vocal folds - Because speech is loaded with redundant cues - Redundancy - multiple ways of conveying the same linguistic category (syllables, words, phrases) Redundant cues for phonological voicing - Initial consonants - Consonant duration is a powerful cur - Stops - Voice onset time - the time from the release of the burst to the onset of vocal fold vibration - Longer for voiceless aspirated stop consonants - Fricatives - Voiceless fricative are longer than voiced fricatives  - Consonants following a vowel - Vowel duration is a powerful cue - A vowel before a voiced consonant is longer than a vowel before a voiceless consonant - “The use” versus “to use” - Lengthening the vowel in the word “use” as in “the use” will cause listeners to perceive the word “use” as in “to use” even if the final fricative is not voiced - Vowel before voiced consonant - longer - Vowel before voiceless consonant - shorter Clinical applications of phonological voicing - How does phonological voicing help us clinically - Durational cues can “mark” intended voicing - We can tell if they intended to voice or not - When present they indicate a speaker understand the phoneme is voiced even when no physiological voicing is present 3 types of stop consonants 1. Voiced stop consonants a. /b/, /d/, and /g/ b. Sequence of events i. Simultaneously 1. The vocal folds fully adduct 2. The soft palate raises and closes off the nasal passages 3. A closure is made in the oral cavity ii. The vocal folds vibrate during the closure 1. The larynx drops, creating a slightly lower pressure above the vocal folds 2. Air moves through the adducted vocal folds, causing them to vibrate until the pressure above and below the vocal folds are equal iii. The oral closure is released 1. Creating a burst - very short noise - also called an impulse iv. The vocal folds vibrate for the following vowel - there is NO ASPIRATION (turbulent airflow generated by approximated vocal folds) 1. If the vocal folds were able to vibrate through the entire stop phase, they will continue to vibrate after the burst 2. If they were not able to vibrate through the entire stop, they will begin vibrating again shortly after the burst c. Voice onset time i. VOT is the time from the release of the burst to onset of voicing for the following vowel ii. For voiced stop consonants, VOT is very short or negative 2. Voiceless unaspirated stop consonants a. Phonemes when in a cluster only  i. /p/ as in “spot”, /t/ as in “stop”, etc. b. Sequence of events i. Simultaneously 1. The vocal folds partially adduct, but are still somewhat open 2. The soft palate raises and closes off the nasal passages 3. A closure is made in the oral cavity ii. There is no vocal fold vibration during the closure iii. The oral closure is released - created a burst iv. The vocal folds immediately close - meaning NO ASPIRATION v. Once full adducts, the vocal folds begin to vibrate for the following vowel c. Voice onset time i. Short, but not negative 3. Voiceless aspirated stop consonants a. Phonemes when not in a cluster i. /p/ in “pot”, /t/ in “top”, etc. b. Sequence of events i. Simultaneously 1. The vocal folds are fully abducted 2. The soft palate raises and closes off the nasal passages 3. A closure is made in the oral cavity 4. There is no vocal fold vibration during the closure ii. The oral closure is released - creating a burst iii. The vocal folds begin to close iv. As the vocal folds are closing they generate turbulent noise called aspiration v. Once fully adducted, the vocal folds begin to vibrate for the following vowel c. Voice onset time i. Voiceless aspirated stop consonants have a long VOT 1. Because the vocal folds are still closing after the burst there is a delay in voicing and aspiration noise occurs Phonology of the 3 types of stop consonants - Voicing of stop consonants is phonemic - Changing the voicing changes the meaning of the morpheme - Aspiration of stop consonants is allophonic - Substituting an unaspirated voiceless stop for an aspirated voiceless stop does not change the meaning of the morpheme Perception of English stop consonants - 2 stops with short voice onset times (VOT - Voice stop consonants - very short, or negative - Unaspirated voiceless stop consonants - short, not negative - When an unaspirated voiceless stop consonant is produced in an atypical place (ex: not in a cluster), the short VOT causes it to be perceived as a voiced stop consonants Additional voicing term - Cognates - paired of voiced and voiceless stops, fricatives (except h), and affricates - Many speech sounds occur in pairs which only differ in voicing - /p,b/, /f,v/, /θ,ð/, /t,d/, /s,z/, /ʃ,ʒ/, /ʧ,ʤ/, /k,g/ Context limited phonemes - English places some limits on where particular sounds can be used - /h, w, j/ never word-final - /ʒ, ŋ/ never word initial - Lax vowels - never in stressed, open syllable - /i,æ/ almost never before /r/ - /ɚ, ə, e, o/ only in unstressed syllables - /ɝ, e͡ɪ, o͡ʊ/ only in stressed syllables - /ɾ/ only in intervocalic position, before an unstressed vowel SUPRASEGMENTALS What are suprasegmentals - Individual speech sounds are sometimes called “segments” - Speech involved more than just stringing together individual segments in a sequence - Other aspects of speech are laid on top of groups of segments - Called the suprasegmentals Divisions of suprasegmentals 1. Prosody a. Aspects of pitch, loudness, and duration that occur on several levels of speech i. Sentence, word ii. Includes intonation, tempo, and loudness 2. Paralinguistics a. Expression of voice quality, emotion, and speaking styles b. These are what motivates prosody 3. Coarticulation PROSODY: Intonation - includes several related phenomena - Rhythm - how stress is distributed - Enabled listeners to predict upcoming information - makes for more efficient listening and communication overall - Helps us deal with “noisy” situations - Helps us predict when we have out turn in the conversation - Languages tend to fall into three groups based on rhythm - Stress timed languages - stressed syllables occur at roughly equal intervals of time - Tend to see short rushes of speech as the time for the next stressed syllable approaches  - Ex: Germanic languages such as English, German, or Dutch - Syllable timed languages - a fairly consistent number of syllables occur every unit of time - rate is likely to be more consistent within a speaking turn - Ex: Romance languages, such a French, Italian or Spanish - Mora-timed languages - a consistent number of “mora” or weight-units occur every unit of time - Ex: Japanese - Weak or unstressed syllables = 1 mora - Stressed syllables = 2 mora - Pitch declination - pitch usually drops slowly over the course of a clause or sentence - Provides clues as to where the boundaries of language units are, such as phrases or clauses - New vs. given information - when new information is being introduced, speakers often use higher pitch or longer duration - Contrastive stress - highly related to new information - sometimes certain words or phrases need extra emphasis to be sure the specific meaning is being conveyed - Usually trying to contradict something previously spoken with opposite meaning - Usually done by stressing those particular words or phrases - Lexical stress - stress within a word can sometimes make all the difference in meaning - There are many instances of words in which stressing a different syllable totally changes the meaning - Ex: contract, digest, increase, object, content - Stress can also help us distinguish between compound words and phrases - Ex: White House vs. white house - Tone - pitch changes that are applied over a single phoneme or syllable - results in a change in meaning (tone can be a phonemic feature) - Occurs in many languages, but not in English PROSODY: Tempo - a second major aspect of prosody - Tempo = pace of output - how fast we speak - Three aspects: - Pause (juncture) - periods of momentary silence - Filled pause - using “um” or “like” - At least ⅔ of pauses occur at grammatical boundaries such as the ends of phrases, clauses, or sentences - Most speakers are very skilled at increasing or decreasing their tempo by making their pauses longer or shorter - No absolute definition of what defines a pause - Must account for both normal spacing between words as well as silent events such as to stop phase of stops and affricates - Researchers typically define a pause as any silent period of 200-250 ms or longer - Phrase-final lengthening - sometimes called pre-pausal lengthening  - Tendency to make the last element in a grammatical unit longer than the previous ones - Most noticeable before a well-defined pause - Like pitch declination, it helps listeners identify ends of grammatical units, helps them to “parse” the incoming message - Speaking rate - as we talk faster the duration of individual speech components get shorter - To make the elements shorter, we usually make the movements themselves shorter - Result = undershoot - In fast speech some unstressed or weaker segments may actually be left out - In fast speech, research suggest that the timing and organization of the movements may be different than in normal rate - This means that when we ask speakers to do things very fast, we may not be seeing normal speaking patterns - may not be the best way to assess speech skills PROSODY: loudness - How intense the signal is - perception of intensity - There is no generally accepted way to transcribe loudness - Some symbols have been suggested for analyzing loudness in discourse Bases of prosody - What makes stressed syllables stand out - Most would say it sounds louder - Speakers may make it louder but they also make it longer or higher in pitch - Typically longer, louder, and higher pitch - Regardless of what speaker does, stressing a syllable also results in larger articulator movements Prosody and development - Much research has suggested that infants may make use of prosody before they master the speech segments - “Child directed speech” includes exaggerated prosody - child will learn language better - Appears to help the infant to identify the various units of the language they are hearing - Children may learn speech in the following order - Learn to control fundamental frequency (pitch) - Learn to control timing of control of the larynx and articulators - Learn to coordinate all movements needed for speech sounds (by age 9) - Learn to combine prosody and movements for sounds (by puberty?) Paralinguistics - Emotional state - may significantly affect the prosody of speech - Anger - fundamental frequency, intensity, and tempo often increase greatly  - Fear - fundamental frequency, intensity, and tempo often increase moderately - Grief - fundamental frequency, intensity, and tempo often decrease moderately - Speaking style - we each have our own style of speaking but we usually vary that style to suit the situation - An important distinction for SLPs is - Clear speech vs. conversational speech - We use clear speech when we are not sure how well we are being understood - Our rate is slower - We avoid reducing vowels - Obstruents (stops, fricatives, affricates) are produced with greater intensity - Voice quality - a voice that is excessively hoarse, harsh, breathy, or nasal can interfere with our ability to understand what is being said - Some speakers (especially children) tend to modify the quality of the voice in certain situations - Assuming the role of a character - Physical conditions (ex: upper respiratory infections) may affect voice quality and affect our ability to understand the message Coarticulation - Unless it is produced by itself, the production of any individual speech sound is frequently influenced by the sounds around it - It is likely that the movements and placement for each sound are modified somewhat to make the task more efficient - We see these modifications in the nasalization of vowels near nasal consonants - Examples of coarticulation: - The /s/ in “goose” versus “geese” - In goose the /s/ is produced with lip rounding, in “geese” it is produced with lip spreading DIACRITICS Diacritics - additional markers used on IPA transcriptions in order to give more detail as to what the speaker produced - Broad transcription - when we only use the phoneme symbols - Narrow transcriptions - when we add diacritics to the phoneme symbols, we are now transcribing allophones - Diacritic symbols placed around the phoneme - Sometimes called close transcription - Several systems - we’ve learn Shriberg and Kent’s system - Use of diacritics can establish the difference between a phonological vs. articulation disorder - If needed, diacritics usually indicate that the speaker knows which phoneme to use but they are having difficulty producing it accurately  - May also indicate that the speaker hasn’t figured out where to use particular allophones - In Kent and Shriberg’s system, the diacritics are placed so that related symbols are in the same place relative to the phoneme symbol - Stress symbols above - Tongue change symbols below - Timing and juncture symbols to the right On Glides/Off Glides - Occasionally speakers insert a very brief or partial sound just before (on glide) or just after (off glide) a fully articulates phoneme → an extra sound - Usually the result of sound escaping while the tongue travels to or from the target sound - Most common is an intrusive “schwa” Intrusive sounds - Occasionally sounds will intrude into our pronunciation as a natural outcome of particular sound sequences - These reflect patterns of timing that are affected by co-articulation - By changing the position of the articulators in a particular order, we accidentally created the new (unintended) sound - Not uncommon to see stops intrude following nasals - Ex: “warmth” → /wormθ/ → /wormpθ/ - In anticipation of /θ/ the velopharyngeal port closes before the lips are released for /m/ - Slight pressure builds up behind the lips and when the lips are opened, we hear a slight burst that we perceive as /p/ - Glide consonants may intrude between two vowels - Recall that glides are rapid movements between two vowel positions - When we produce phrases like “see it”, we don’t usually separate the words, resulting in an intrusive /j/ glide (also with “now it”) - These intrusive glides are ignored for transcription purposes unless they stand out as unusual Syllabification - The consonants /m/, /n/, or /l/ will occasionally take on the role of a syllable nucleus and become syllabic consonants - This occurs most frequently when the preceding consonant is homorganic with the syllabic one - If there is a syllabic consonant, no vowel is indicated - Tongue or lip closure is maintained for the preceding consonant - For /m/ and /n/ the velopharyngeal port open, allowing nasal radiation - For /l/, the sides of the tongue drop to allow airflow around the sides of the tongue - When the two consonants are not homorganic, the place of articulation has to change and thus a vowel (almost always schwa) will likely be produced in between - will not have a syllabic consonant Synchronic tie - Different from consonant clusters - Resembles the tie bar for diphthongs - is used when two sounds are produced together as a single unit - Ex: /d/ and /z/ to produce /d͡ z/ ; /t/ and /s/ to produce /ts͡ / - This is usually not a normal production - May see it in second language learners or young children Unintelligible - Sometimes during transcription we are unable to figure out what word the speaker was trying to say - Need to indicate this in the transcript - Recall that listeners can reliably count the number of syllables heard - Put a /*/ for every syllable heard that was unintelligible - We can then estimate the number of words that were attempted by dividing the following estimators - 1.2 syllables per word - up to 1st grade - 1.3 syllables per word - up to 9th grade - 1.4 syllables per word - 10th grade and older - Percent understood = # of intelligible words divided by the total # of words (intelligible + unintelligible) Nasal emissions - This is a relatively uncommon but important distortion in which there is an abnormal release of noise energy from the nose - May be the result of damaged velopharyngeal port - Unlike nasality, which is usually associated with vowel, nasal emission is usually associated with consonants - Usually audible; may be accompanied by flaring or grimacing of the nostrils - For children with cleft palate this may occur on many sounds (especially obstruents) - For other children, it may only occur on specific phonemes suggesting it has been learned Whistled - A high pitched whistling sound that accompanies fricatives - Most commonly seen on /s/ and /ʃ/ - It is not clear what exactly speakers do to generate this sound Trilled - Trilled sounds are phonemic in some languages but not English - An articulator, typically the tongue tip, is held in a fixed position and air passing over it sets it vibrating - Commonly heard example is the trilled [r] CLINICAL TRANSCRIPTION 3 basic steps of clinical transcription 1. Obtaining a sample 2. Transcribing and scoring 3. Analyzing the results Obtaining a sample - Recording condition - The quietest environment possible - Position yourself to get a clear view of the speaker’s face (make sure the lighting is adequate) - Recording equipment - Use the best quality equipment you can get - Don’t use your phone - Don’t use built in microphones - Best - lavaliere (clip on) or head mounted types - Keeps distance between mouth and microphone constant - best signal to noise ratio - Table top microphone are ok if recording condition are good (ex: in a sound booth) - Use high quality digital equipment - Save as a wave ile not an MP3 - Be familiar with how the equipment works - Practice before you use it - Use manual volume control - At the beginning of the recording session say the speaker’s name and the date for future reference - Maximizing the sample - Spontaneous conversation = best - Be casual about the presence of the recording equipment - Use a variety of material and topic to keep the speaker talking - Make notes about things you see or hear but may not be noticed on the tape - Ex: facial grimace, facial distortions Transcription and scoring - Many factors affect the results obtained - Client factors - Age - generally more difficult to work with children - May not understand directions well - May move about and change the mouth-microphone distance - May also create distracting noises - May be harder to understand because of incomplete speech development - Dialect  - May be a problem if the speaker uses a dialect that is unfamiliar to the examiner or transcriber - Physical and personality traits - Speakers who use limited jaw movements during speech may be harder to understand - Speakers with reduced or excess nasal resonance are harder to understand - Speakers who are familiar to the listener (transcriber) often easier to understand - May bias our judgements - Intelligibility - how understandable the speaker is overall - We can’t judge how accurately a person is at producing speech sounds if we don’t know what the words are supposed to be - Not uncommon to listen to the recording first just to write down what the speaker intended - Then do transcription as a second step - Task factors - Complexity - System - two way scoring, five way scoring, phonetic transcription - Linguistic - single words, sentences, conversations - Response - focus on one sound or all sounds being produced - Successive judgements - we rarely hear just a single example of a speech sound - What we accept as correct or incorrect may be affected by what we did before - Ex: if we transcribed 5 examples of /s/ as dentalized already, we are more likely to call the next one dentalized - The next example might need to be “hypercorrect” for us to say this it was correct - Likewise if we’ve been calling them correct, we are more likely to continue calling them correct - Selecting a scoring system - Although many clinicians use two-way or five-way scoring, the best way to use these systems is really to understand the phonetic transcription first - Gives you solid reasons for saying something is correct or incorrect or a substitutions vs. a distortion - Selecting a symbol system - Use phonetic symbols for transcription - Need alternatives for two-way and five-way scoring - Each clinician usually decides what works best for them as long as they are consistent - Five way - five possible types of response - Correct, omission, substitution, distortion, addition - Could abbreviate each response - For substitutions and additions we often also indicate what was new  - Ex: [t/s] - symbol on left is what was produced, symbol on right is the target - Selecting a recording form - Most published “articulation” (could be phonological or articulation based) tests have their own specifically designed form - For two-way scoring, many clinicians keep a tally of correct and incorrect for the sounds they are listening for - For transcription, need 3 levels of information - What the speaker intended to say X LINE - Use orthography - sometimes called gloss - The accepted phonetic target Y LINE - What normal adult speakers of that dialect would have produced - using phonetic description - What the speaker produced Z LINE - Detailed phonetic transcription of what the speaker actually said - Defining responses - With two way or five way scoring, we need to know what we mean by correct or incorrect - Need to know what to do about ambiguous production (those that could go either way) - Could ignore (but keep a tally) - Could assume they are correct - Don’t want to dismiss a client from therapy too early - Could assume they are incorrect - Don’t want to penalize speaker for a problem that is not really there - What to transcribe/ignore - For most clinical situations, we would not indicate all normally produced allophones - Ex: aspiration of word-initial stops - Linguists may need this level of detail Analyzing the results - For two way scoring, count number of errors - May also calculate percent correct or incorrect out of total attempts - Five way scoring - Count number of each types of error - May also calculate percent of each error type out of total attempts - With published tests the number of error is usually then compared against some “norms” or “standards” to determine if there is a problem - For transcription we compare the Y line with the Z line - We then calculate the percent of correct or incorrect either overall or for each sound type Considerations for connected speech transcription - Spaces between word are artificial - We produce connected speech in larger units (more than just one word at a time) - Spaces added to make analysis easier - Several changes more likely (especially in casual speech) - More vowel reduction - More intrusive consonants - More coarticulation leading to loss of some elements CLINICAL SOUND CHANGES Vowel/diphthong changes stop consonant changes Fricative and affricate changes Liquids and glides changes Nasal changes Each class of sounds is subject to particular kinds of changes Vowel/diphthong changes - Lip rounding/unrounding - English contain a mix of both rounded and unrounded vowels - Some speakers may add lip rounding to normally unrounded vowels or not round their lips for normally rounded bowels - May reflect problems with neuromuscular control of the lips - Ex: lip spasms - Young children who delete final consonants (which involve the lips) may add lip rounding to a normally unrounded vowel that occurs before the consonant - This is termed marking - This suggests that they are aware on some level of the need for the consonant but may still be unable to fully produce it - In other words, they know it is there phonemically but they can’t make it yet - Would need artic therapy not phonological therapy - Lip Rounding changes are hard to identify over audio recordings alone - Vowel substitutions - Typically the new vowel is from an adjacent position in the mouth - Many dialect differences (English) are seen as normal vowel substitutions - Spanish differs more on consonants - On glides/off glides vs. diphthongization - A vowel on glide or off glide added to a vowel can sound similar to diphthongization - Difference is only a matter of degree - Id one of the vowels is clearly brief and unstressed, transcribe it as an on glide or off glide  - If both about equal, transcribe as diphthong - Don’t use tie bar if you clearly hear two separate syllables - Vowel lengthening - Sometimes vowels are extended in duration - Indicated by : to the right of the phoneme - Somewhat subjective but a lengthened vowel will stand out as unusually long - Difference can be important clinically - Vowel are normally somewhat longer before voiced consonants - This is a normal allophonic difference - Marking - Many children delete final consonants - We need to listen for the length of the vowel before the missing consonant - If they only lengthen it when the missing final consonant was supposed to be a voiced obstruent, they are telling us that they know the consonant is supposed to be there - They are marking it - In southern American English, diphthongs become monophthongized - but lengthened - It is lengthened to mark the difference between word pairs such as tar/tire, knot/night - Nasalization - Happens on vowels - Nasality can be phonemic in some languages - Normal to add some nasal quality to vowels adjacent to nasal consonants - This is known as assimilative nasality - If the vowel before a missing nasal is nasalized the child is marking the missing nasal Stop consonant changes - Substitutions - /p/ and /b/ are least likely to be in error - because they do not require tongue movement - May see it in children with cerebral palsy - For alveolar and velar stops, substitutions usually involves change in place of articulation - Rare to see a change in manner - Commonplace errors = fronting [t/k], or backing [k/t] - Not that voicing is usually the same - Aspirated - Voiceless stops at the beginning of syllables are usually accompanied by an audible emission of air produced by closely approximated vocal folds (aspiration) - Not normally heard with voiced stops - Indicated by a superscript “h” to the right of the phoneme - Unaspirated - In the second position in a cluster, no audible emission of air occurs  - Indicate “unaspirated” by a superscript “=” to the right of the phoneme - Unreleased - Occasionally, at the end of an utterance, there may be no release phase - Indicate unreleased by a superscript right angle to the right of the phoneme - Aspirated, unaspirated, unreleased, are normal english allophones - Not normally shown on a transcript unless they stand out as unusual - Voicing changes - When voicing is fully lost or fully added, we get a complete change in the phoneme - It is also possible to have partial voicing changes - Two common partial changes seen - Initial voiceless stops may appear to be partially voiced - To our ears, this case sound very similar to an unaspirated voiceless stop - But partial devoicing also sounds very similar to devoicing of the cognate sound - Have to be specific when transcribing - substitutions are considered more severe than distortions) - Final voiced stops become devoiced - Quite common even for normal adult speakers - Listen for the duration of the vowel - Glottal stop substitutions - Hard to distinguish from normal onset of voicing - Children do this mostly in final position - Children with cleft palate or other velopharyngeal mechanism problems may use glottal stops in all positions - In initial position, it sounds like the vowel starts very abruptly - In final position, it sounds like the vowels end very suddenly - Stop deletions - Two common situations occur regarding stop deletions - Unreleased stops may be perceived as deleted, especially is certain contexts - When another stop follows it - When it is between two other consonants - In a final homorganic cluster - Very young children or those with delayed speech often delete stops at the ends of words - Called final consonant deletion - Watch for visual cure to see if the stops were unreleased in word final position - Pay attention to the preceding vowel - If lengthened or nasalized or if there is an off glide, child may be marking missing sound - Frictionalized stops - Aka spirantized - Burst is less well defined - stop is “noisy  - Result of a slow gradual release of the stop closure - Dentalized stops - Only applies to /t/ and /d/ - Tongue tip created closure near or at the lower edge of the upper teeth - May be used in place of correct /t, d/ but may also be substitutions for /θ,ð/ Fricative and affricate changes - Continuants - sounds that can be prolonged without interruption - Includes fricative, liquids, glides, and nasals - Vowels are also continuants - Lisp - popular press uses this term to mean any speech error - Actually refers to distortions of fricative and affricates - May be seen as derogatory to some - A lateralized /ʃ/ or /s/ is sometimes called a lateral lisp - Fricatives are frequently changed - Voiceless fricatives are usually produced with greater intensity and longer duration than the voiced fricatives - This means errors on voiceless fricatives are more likely to stand out - Types of changes - Deletions - Speakers who speak softly may appear on tape to be deleting fricatives - Non sibilants are weak sounds - Easily lost in recordings - Weakly produced sibilants may be confused with recording noise - In casual speech /h/ is frequently omitted in unstressed positions - /θ/ and /ð/ often omitted in rapid speech or consonant clusters - Stopping - Very young children or those with delayed speech often substitute stop consonants for fricatives - Usually but not always the stop and the fricative are homorganic or almost so - The voicing feature is usually retained - Thought to occur because production of stops is believed to be similar than production of fricatives - Affricates may be reduced in speech delayed children to stops - Other substitutions - Fricatives may be substituted for each other - With affricates, stops may be substituted - The fricative portion may be changed - Voicing changes - Fricative and affricates follow the same patterns as stops - Voiceless forms may be partially voiced and voiced forms may be partially devoiced - Other distortions - Dentalization (frontal lisp) - Usually only apply to the sibilants - occasionally the affricates  - The tongue tip is moved forward and may contact the alveolar ridge or teeth - Tongue groove is maintained or even tightened - Sounds part way between /s/ and /θ/ - Most common distortion error is dentalized /s/ - Lateralized (lateral lisp) - May occur on all the sibilants - Uncommon but difficult to remediate error - Air is escaping around the sides of the tongue (which normally create a seal against the upper teeth) - Retroflexed - Relatively uncommon error occurs on all sibilants - Tongue assumes a retroflex position as if it was producing an /r/, but a sibilant is generated - Glide substitutions - Glides are among the first phonemes to be learned - Errors on glides would indicate extremely delayed speech - If they occur, the most common errors would be to substitute glides for each other - Liquid substitutions - Liquids are among the more difficult sounds to produced - /l/ is often replaced b /w/ or /j/ in initial position - In final position /l/ may be replaced by a vowel, especially back vowels such as /o/ - A common error in very young normal children is that /r/ is replaced by /w/ and occasionally by /j/ - Also seen in older speech delayed children - Liquid /l/ distortions - Many speakers produce a dark or velarized /l/ in final position - The tongue tip is pushed back along the palate aways from the alveolar ridge - Makes it sound more like a back vowel - Some speakers use /s/ in all positions - Liquid /r/ distortions - Both /l/ and /r/ may be velarized, though it is a rare error with /r/ - All forms of /r/ may be produced without the rhotic quality - They are derhotacized /r/ - Not truly /r/ and not truly any other phoneme Nasal changes - Nasals are early acquired sounds - Omissions - With very young children, omission of nasals may be hard to notice because of the assimilative nasality that we hear on the adjacent vowel - Common substitution of n/ŋ at the end of present progressive verbs like “running” or “jumping” usually not seen as an error  - Especially if the syllable is unstressed - Substitutions - Some young children confuse nasals and substitute one for another - Devoicing - Nasals may be devoiced when the /s/ is omitted from /s/ + nasal clusters - Denasalization - A head cold or some structural problem that prevents nasal airflow, may affect nasal sounds - May be replaced by its homorganic voiced stop - May be followed by an intrusive off glide homorganic voiced stop - In both cases the adjacent bowels will be denasalized DIALECTS Dialect - a variety of language - Generally two individuals who speak different dialects of the same language should be able to understand each other - Mutually intelligible but distinct - Dialects share more features than they differ on - Most dialects tend to evolve when groups of people are geographically and/or socially isolate from other groups - Speakers of dialects of the same language that are the most isolated from each other tend to be the least able to understand each other - May eventually evolve into separate languages - All languages that are used undergo change over time - When groups are isolated from each other, different changes may occur within each group - Result → different dialect - Over very long periods, it is believed that dialects eventually evolve into separate languages The standard or regional dialects - The dialects of American English that most people think about tend to be geographically defined - May be called the standard dialects because they represent the regional standard - Initially the regional dialects of the US reflected the settlement patterns of immigrants from various regions of Britain - Changes in the dialect occurred over time in each region to bring each to where it is today The social dialects - The dialects of American English spoken by specific social groups - In every region there is a mainstream dialect and a vernacular dialects - Don’t usually have a more prestigious dialect  - The vernacular dialects are sometimes referred to as non-standard dialects and are often seen as less prestigious Dialect differences - Probably see differences in all aspects of language, but especially in - Phonology - biggest group of differences found in the sound system. Differences = accents - Semantics - often many differences among dialects in particular words chosen - morphology/syntax - some differences seen in word and/or sentence structure Origin of dialect differences - May result from tendency to try to make language easier to use and learn - Generalization of rules - Existing rules are extended to other contexts - Ex: “knew” → “knowed” - Irregular forms become more regular - Redundancy reduction - Delete features that are already indicated - Ex: “he likes” → “he like” - Naturalness - Sounds are changed or omitted so the production is either easier to hear or produce - May also reflect system-wide changes - Ex: if one sound is changes, the result may be the loss of contrast with an adjacent sound - In order to retain the contrast in meaning, other changes may have been added - A chain shift - to maintain the contrast in meaning among a group of words all of them change - A series of changes are linked together - May also reflect the creation of simpler syllable structures - Ex: nuclear → nucler - Cvccvv → cvccvcv American English Dialects - Shriberg and Kent say they are teaching us about General American English (GAE) aka standard american english - SAE is a less desirable term because it assumes there is one standard when there is not - GAE = version of English that is “homogenized” - Cannot be easily recognized as coming from a particular region of the country - Used by many newscasters and is sometime called “Network Standard English”  - The regional or standard varieties (dialect) of American English all include some degree of change from GAE - The changes seen in a particular dialect are not necessarily limited to speakers of that dialect - Ex: some dialects share features - It is the collection of differences that makes a particular dialect unique Black English (AAVE) - The following terms all mean the same - African American English - African American Vernacular English - Black English - Black English Vernacular - Black dialect - Ebonics - In 20th century common dialect for poor and inner city neighborhoods - Meaning not everyone who uses it is African American - May have arisen in the same way as the standard varieties did (from British dialects) - Called anglicist hypothesis - Many sociolinguists believe it arose via a pidgin-creole process - Called the creole hypothesis Creole Languages - Native Pidgin language - Two most common creole’s in the US - Louisiana French Creole (cajun) - Contact between Acadian French, native americans, and southern american english speakers - Word cajun - condensed form of Acadian - AAVE (hypothesis) - If true it was developed by African slaves who originally spoke many different languages in Africa - Slave traders often mixed slaves from different countries to avoid mutiny on the voyage to America or Europe - The pidgin language the slaves developed among themselves was later heavily influenced by the English used by the slave masters AAVE features - Phonemic diphthongs reduced to monophthongs - /ŋ/ → /n/ in unstressed syllables as in the present progressive-ing - Frequency omission of final single consonants especially under particular conditions - If final C = stop or nasal - If next word begins with a consonant - If sound would add redundant meaning - If part of a cluster that does nor add another morpheme  - Loss of one or more elements from many word final consonant clusters - Dental fricatives /θ, ð/ mostly absent (changed from alveolar stops or labiodental fricatives) may see /θ/ in initial position - Syllable final /r/ becomes vowelized - /v/ → /b/ when a nasal occurs later in the word The logic of non-standard English - Although some dialects are said to be more prestigious than others, no dialect is any more or any less logical than any other dialect - all dialects are rule-governed and conventionalized (speakers agree on rules) - Speakers of upper social classes tend to want to maintain the status quo - May be more resistant to change - Speakers of lower social class are less resistant to change - More likely to adopt the previously discussed natural changes - Speakers of lower social class also tend to be speakers of non-standard dialects - The non-standard dialect would be more likely to show these changes than the standard dialects - Recall that the changes are more regular (fewer exceptions), less redundant, and possible more natural - If the non-standard dialects use them more often, it could be argues that the non-standard dialects are more logical than the standard dialects - It is noteworthy that many of these changes so eventually tend to be adopted by the standard dialects as well - Far fewer irregular past tense verbs today than 1000 years ago Code switching - Because of the diversity of US society, many individuals are bi-cultural and bi-dialectal - More likely with members of minority groups or speakers of non-standard dialects - Many of these individuals learn to interact in more than one cultural or dialect setting - They figure out the rules for each and learn to use one set of rules in each setting - Some school districts actively teach children how to code switch Style switching - Speakers of the same dialect may use different styles with different audiences - Reflects differences in level of formality which means differences in vocabulary, syntax, and vocabulary - We all do this Slang vs non-standard dialect - Not the same thing - Slang has 3 unique properties - Only used in informal situation - Tends to be restricted to a particular age group - Meaning only understood by that group - Has a very short life span