Language & the Mind Sound LING240 Summer Session II 2005 Production Lecture 5 Sounds How you look to a phonetician How you look to a phonetician Nasal Cavity Palate Velum Oral Cavity Tongue Glottis (vocal folds) Lips, teeth etc. T ‘thick’ D ‘the’ S ‘sh’ Forget Spelling! Z ‘azure’ Sounds 䍫 Spelling tS ‘ch’ dZ ‘j’ N ‘thing’ / ‘ uh- oh’ 1 One Sound - Many Characters One Sound - Many Characters he e seas ea too oo threw ew believe ie amoeba oe to o lieu ieu Caesar ae key ey see ee machine i clue ue shoe oe people eo seize ei through ough Interantioanl Phonetic Alphabet: [i] IPA: [u] One Character - Many Sounds One Sound - Multiple Letters dame e shoot S dad Q either D character father a k deal call ç i Thomas t village I, ´ physics f many E rough f One Letter - 0, 1, 2 Sounds Differences across Languages mnemonic • English: judge, juvenile, Jesus [dZ] psychology resign • Spanish: jugar, Jesus [h] ghost • German: Jugend, jubeln, Jesus [j] island whole • French: Jean, j’accuse, jambon [Z] debt cute [kjuwt] 2 Major division: consonants vs vowels • Consonantal sounds: narrow or complete closure somewhere in the vocal tract. • Vowels: very little obstruction in the vocal tract. Can form the basis of syllables (also possible for some consonants). Describing Speech Sounds Where does the Air Flow? • Where/how is the air flowing? nasal/oral, stop, fricative, liquid etc. • Where is the air-flow blocked? labial, alveolar, palatal, velar etc. • What are the vocal folds doing? voiced vs. voiceless Block it at the velum Where does the air go? Your vocal tract again 3 N Block it at the velum Tongue against velum again Where does the air go? Now raise the velum Now raise the velum to block the air.... to block the air.... Quickly drop your Quickly drop your tongue again ... tongue again ... 4 Where does the air Where does the air go this time? go this time? So far we have: g Nasal stop: Where does the air [N] go this time? k Non-nasal (oral) stops: [g] [k] ? ? m m a a e e r r t t s s r r i i a a e e h h t t p p o o t t s s u u o o y y n n a a c c (bi)labial e e r r [b] [p] [m] e e h h W W 5 Where can you stop the airstream? Where can you stop the airstream? Where can you stop the airstream? [ v [ a [ l a g d v l e v b ] ] ] l e i a [ [ [ o o r k t f d l ] ] a ] e [ r n [ n N t ] a ] l [ s ] [ z ] [ l ] [ r ] Where can you stop the airstream? Where can you stop the airstream? Where can you stop the airstream? [ p [ i u n D Z a v t ] ] l u e a [ [ r l t a T S d a °] r ] e l n [ d t a Z l ] [ t S ] ] 6 ? m a e r Manner - How the Air is Flowing t s r i a e • Stops h t [p] [t] [k] [b] [d] [g]... p o t • Fricatives s u [f] [v] [T] [D] [s] [z] o y • Approximants/Glides n a [w] [j] c laryngeal e • Liquids r e [r] [l] h W Fricatives & Affricates Voiced & Voiceless Consonants • Palatal sounds [Z] [S] [dZ] [tS] • Consonants either voiced or voiceless. • English pairs: • Palatal Fricatives - [Z] [S] [note: according to IPA chart these are strictly ‘post-alveolar’] b p v f d t • Affricates - combination of stop + z s D T fricative - [dZ] [tS] , as in judge, church Describing Consonants Features • Where is the air-flow blocked? • Ways of describing sounds labial, alveolar, palatal, velar etc. e.g., [t] = voiceless, alveolar, stop • Where/how is the air flowing? • Stronger claim: features are the nasal/oral, stop, fricative, liquid etc. smallest building blocks of language, • What are the vocal folds doing? used to store sounds in the mind voiced vs. voiceless • Atoms of Speech Roman Jakobson, 1896-1982 7 Features • Prediction: by combining a small number of atomic features, it should be possible to create a larger number of speech sounds • Goal: a set of universal features should make it possible to describe the speech sounds of all of the languages of the world • Different languages choose different feature combinations ? ? “año” ? “Fuji” “Cuba” ? ? “Bach” “agua” ? ? ? ? “caballo” 8 What can you do to alter the shape of your vocal tract? [i] [Q] 9 [I] [u] So what vowels do you have? You can.... i “sheep, sleep” • Raise or lower your tongue I “ship, slip” • Advance or retract your tongue • Round or spread your lips • Tense or not tense your mouth So what vowels do you have? So what vowels do you have? i i I I “laid, spade, trade” eE “led, sped, tread” eE Q “bat, lad” 10 So what vowels do you have? So what vowels do you have? i “Luke, who’d, suit” u i u I “look, hood, soot”U I U “coat, wrote, hoed” e e E E ç o “caught, wrought, hawed” Q Q So what vowels do you have? So what vowels do you have? i u i u I U I U e o e o E ç E“but, putt, rut” ç √ “bah, father, cot, Don” Q a Q a So what vowels do you have? So here they are! i u i u I U I U “metallic, Texas” e e E ´ ç o E ´ ç o √ √ Q a Q a 11 Some dialectal differences Cross-language Differences • Feature Combinations • caught/cot [Mid back lax vowel and mid – English: back vowels are rounded, others are not back tense vowel]: many American – German/French has high, front, rounded vowel [y] speakers do not have both of these. – Russian has high back unrounded vowel [µ] • pot/father: some British and (fewer) American dialects have different vowels – Many languages don’t make the tense/lax in these words (“pot” has a low back distinction found in English (ex: Spanish [i]) – Many languages distinguish short and long vowels rounded vowel [Å]). (unlike English), ex: Japanese Diphthongs: Cross-language Differences a Diphthongs: Diphthongs: “side, my, kind” aj a 12 Diphthongs: Diphthongs: “loud, brow, hour” ç aw Diphthongs: Speech Production - Summary • Airflow set in vibration by vocal folds Airflow modified by vocal tract “boy, annoy, toil” • Vowels: shaping of oral cavity • Consonants: narrowing or blocking of çj oral/nasal cavity • Different languages choose different selections of articulatory gestures Speech Perception Acoustic Information • Speech production processes must be • Frequency undone by the ear • Motions of articulators must be reconstructed from patterns of air • Timing vibration • Requires extremely precise hearing, possibly a system specialized for hearing speech • Substantially developed at birth 13 Frequency - Tones Frequency - Tones (Close Up) Frequency - Tones Frequency - Vowels • Vowels combine acoustic energy at a number of different frequencies • Different vowels ([a], [i], [u] etc.) contain acoustic energy at different frequencies • Listeners must perform a ‘frequency analysis’ of vowels in order to identify them (Fourier Analysis) Frequency - Male Vowels Frequency - Male Vowels (Close Up) 14 Frequency - Female Vowels Frequency - Female Vowels (Close Up) Synthesized Speech Timing - Voicing •Allows for precise control of sounds •Valuable tool for investigating perception Voice Onset Time (VOT) English VOT production • Not uniform • 2 categories 60 msec 15 Perceiving VOT Discrimination Same/Different 0ms 60ms Same/Different Why is this pair difficult? 0ms 10ms (i) Acoustically similar? (ii) Same Category? ‘Categorical Perception’ Same/Different 40ms 40ms Discrimination Cross-language Differences A More Systematic Test Same/Different 0ms 60ms D 0ms 20ms D R L Same/Different D 20ms 40ms T 0ms 10ms T 40ms 60ms T Same/Different Within-Category Discrimination is Hard R L 40ms 40ms Cross-Language Differences Cross-Language Differences English vs. English vs. Hindi Japanese R-L alveolar [d] ? retroflex [D] 16 Cross-language Differences Japanese Syllables Participants: Thai – native English- second (>3 years in the US) • English Pronunciation [d1a] [d2a] DIFFERENT [m Q k d a n ´ l d z] • Japanese Pronunciation [m Q k u d o n a r u d o] Imsri & Idsardi (2001) What’s a Syllable? Phonotactic Constraints: Constraints on Syllable Structure • Another phonological unit of words • Every vowel is at the center of a syllable • Every language has restrictions on • Syllables have hierarchical structure what sequences of phonemes may occur (*ktleem) σ • These constraints are language specific English: *zleem Polish: zlev ‘sink’ onset rime • Illegal sequences are illegal within a single syllable nucleus coda English: *[kspl] [Ek splI sIt] s p l I n t s Japanese Phonotactic Constraints Japanese Syllable Structure σ • Toyota onset rime nucleus coda (C) • Honda Japanese doesn’t allow consonant clusters V (n) within a syllable! Toyota, Honda… 17 Japanese Syllable Structure Japanese Syllable Structure Japanese Syllable Structure Japanese Syllable Structure Japanese Syllable Structure Japanese Syllable Structure 18 Japanese Syllable Structure Japanese Syllable Structure Japanese Syllable Structure Behavioral Results • Japanese speakers have trouble hearing the difference [ebuzo] Phonemic Level: /m Q k d o n a l d/ Phonetic Level: [m Q k u d o n a l u d o] Dupoux et al. 1999 Additional Findings Speech Perception It seems that a language speaker is a prisoner of his/her language phonemic alphabet 19 A Puzzle… Developmental Questions • Korean speakers use the sounds [r] and [l] • How does the native/non-native e.g. Korea difference emerge? Seoul • Korean babies hear the difference between [ra] and [la] … they don’t know Korean yet • Does native-language discrimination • Korean adults know Korean … but they have improve as a result of native language difficulty hearing the [ra] vs.