The Phonetics and Phonology of Nasal Gestures .


18 views
Uploaded on:
Description
The Phonetics and Phonology of Nasal Gestures Patrice Speeter Beddor University of Michigan Supported by NSF. In collaboration with:. Anthony Brasher Rose Letsholo Susan Lin Chandan Narayan Chutamanee Onsuwan Samantha Sefton. Background.
Transcripts
Slide 1

The Phonetics and Phonology of Nasal Gestures Patrice Speeter Beddor University of Michigan Supported by NSF

Slide 2

in a joint effort with: Anthony Brasher Rose Letsholo Susan Lin Chandan Narayan Chutamanee Onsuwan Samantha Sefton

Slide 3

Background The transient and spatial degree of coarticulatory vowel nasalization is known to be exceptionally factor in ways that rely on upon phonetic setting and prosodic structure, including, among numerous different components:  obstruent voicing in NC bunches (Malécot, 1960)  frication in NC groups (Ohala & Busà, 1995; Busà, 2007)  vowel length (Whalen & Beddor, 1989)  syllable structure (Krakow, 1989; Cohn, 1990; Solé, 1995)  stretch (Vaissière, 1988; Krakow, 1993)

Slide 4

Hypothesis: The worldly and spatial degree of the nasal signal is generally consistent over a scope of relevant conditions, in spite of the fact that the fleeting arrangement of nasal and oral motions can contrast crosswise over settings in unsurprising ways. Three-pronged approach: creation, discernment, phonology Central theory For quite a long while, cross-dialect explore on coarticulatory variety in our lab has been persuaded by the desire that the nasal (brought down velum) signal is more steady than might show up from the vowel nasalization writing.

Slide 5

~ Perception: Listeners regard N and V as perceptually proportional and are more touchy to aggregate nasalization in the rhyme than to the exact arrangement of oral and nasal motions. ~ Phonology: The co-variety amongst V and N underway and discernment is predictable with broadly bore witness to phonological designs . Focal theories Production: In tautosyllabic VN arrangements, length of coarticulatory V nasalization and term of N co-shift , with the end goal that velum bringing down in the syllable rhyme is moderately steady crosswise over phonetic settings. Explored for 2 VN settings: taking after C voicing going before V length

Slide 6

Lowered velum Vowel Oral conclusion V N C Same-sized velum motion started before V N C More broad V nasalization Longer post-N oral choking Shorter N mumble Variation in worldly arrangement

Slide 7

English Stimuli:/C E nC/words where coda C =/s t d z/ (e.g., twisted, twist, sent, send, sense, thick, nooks ). Speakers: 5 local speakers of American English. Acoustic measures : N term, post-N oral conclusion span, and length of V nasalization. In measuring V nasalization, FFT spectra were reviewed in 10 ms increases over the V. Nasalization onset = first range with an identifiable low-recurrence FN as well as a widening of F1 BW and bringing down of F1 abundancy. Variety in worldly arrangement: impacts of voicing This example is correctly what we find in voiceless (in respect to voiced) settings in English.

Slide 8

(a) (b) (c) Relative Amplitude 1800 0 (f) (d) (e) Frequency (in Hz) Measuring V Nasalization

Slide 9

V Duration (ms) ~ N Duration (ms) Variation in fleeting arrangement: impacts of voicing AMERICAN ENGLISH Scatterplot of VNC voiceless (filled) and VNC voiced (unfilled) tokens from 5 speakers. Information demonstrate converse connection amongst N and V lengths . (R 2 measurements for pattern lines run from .27 to .45.)

Slide 10

Voiceless Post-N Closure (ms) Speaker 2 Speaker 1 Variation in transient arrangement: impacts of voicing Same-sized velum motion started before N V C More broad V nasalization Longer post-N oral choking Shorter N mumble

Slide 11

Variation in worldly arrangement: impacts of voicing Later onset of the velum signal in voiced settings is likely because of streamlined variables, with nasal spillage before full velum conclusion and the kept raising of the velum after conclusion (which extends the oral hole) encouraging voicing (Hayes & Stivers, 2000; Solé, 2007). Be that as it may, not all dialects show N shortening in pre-voiceless settings. Our expectation is that these dialects ought to likewise not demonstrate more broad vowel nasalization in these specific situations. Ikalanga VNC: As in other Bantu dialects, NC arrangements are generally examined as prenasalized N C (Mathangwane, 1999) . Most NCs in Ikalanga are voiced, in spite of the fact that NC [voiceless] happen in borrowed–including some common–words. Jolts: Extensive word list with voiced VNCV (e.g., [da n da]) and voiceless VNCV (e.g., [ke n ta]) successions Speakers: 6 local speakers of Botswanan Ikalanga

Slide 12

No huge impact of voicing on span of N or coarticulatory nasalization; add up to length of nasalization (V and N) is consistent crosswise over settings. ~ Variation in transient arrangement: impacts of voicing IKALANGA Duration (ms) Duration of oral and nasalized V, N, and C in VNCV groupings.

Slide 13

Lowered velum Same-sized velum signal Shorter oral tightening More broad V nasalization Variation in worldly arrangement: impacts of vowel length A second setting known to offer ascent to variety in N term is length of a first V. Oral conclusion Vowel Short V + Long N V N Long V + Short N V N

Slide 14

Variation in transient arrangement: impacts of vowel length This example is maintained by information from Thai and English. Thai (Onsuwan, 2005) Stimuli: Extensive arrangement of CVN and CV:N words created by 4 speakers American English Stimuli: CVN words with (long) tense and (short) remiss Vs (e.g., seen, sin, torment, pen ) delivered by 6 speakers

Slide 15

Variation in worldly arrangement: impacts of vowel length Duration of N and oral and nasalized V in Thai and English CVN groupings. (Thai information from Onsuwan, 2005) The more N is, the less broad is the expectant nasalization on V.

Slide 16

Variation in worldly arrangement The opposite connection between span of N and length of V nasalization is found for different settings and in different dialects also: Voicing VNT (shorter N, more prominent V NAS) versus VND: English (our lab; Cohn) Vowel length V:N (shorter N, more prominent V NAS) versus VN: English, Thai (our lab) Frication VNS (shorter N, more noteworthy V NAS) versus VNT: Italian (Busà 2007) Japanese (Hattori et al. 1958) Production outline: Coarticulatory vowel nasalization demonstrates extensive worldly variety across settings, yet nasal signals in codas have fleeting security: settings with shorter N—including settings that would seem to instigate N shortening for considerably extraordinary phonetic reasons—have more expectant vowel nasalization.

Slide 17

Variation in worldly arrangement: Perceptual identicalness amongst V and N ~ In perspective of the co-variety underway amongst V and N, for as long as couple of years we have tried the perceptual theory that audience members are more touchy to aggregate nasalization in the rhyme than to the exact arrangement of oral and nasal signals. Our technique utilizes a variation of the exchanging relations worldview in which we orthogonally fluctuate term of N and transient degree of V nasalization. We have tried this approach with speakers of various dialects, and with different sorts of boosts, and I report here one arrangement of discoveries. ~ Stimuli: original [gaba] and [gamba] Using cross-grafting strategies, we co-differed N length with span of vowel nasalization, yielding: (a) 9-stage [b-mb] arrangement which incrementally traded parts of oral beating for the stop with nasal mumble (b) 2 degrees of vowel nasalization: slight (20%) and heavier (52%)

Slide 18

~ Perceptual comparability amongst V and N Stimuli were combined in 3 sorts of separation pairings to figure out if audience members treat nasality on V and N as perceptually identical.  N-just matches: shorter Ns combined with longer Ns (V NAS held steady)  "Different nasality" sets: slight V NAS + short N term paired with heavier V NAS + long N span  "Similar nasality" sets: slight V NAS + long N length paired with heavier V NAS + short N span ~ If heard as perceptually comparable, then:  ought to be hardest for audience members  ought to be most straightforward for audience members Listeners: 28 American English speakers

Slide 19

~ Perceptual comparability amongst V and N ~ g an a m b a g an am b a ~ g an a m b a g an a m b a Different nasality combine ga S m S ba - ga L m L ba Similar nasality match ga L m S ba - ga S m L ba

Slide 20

~ Perceptual proportionality amongst V and N ~ Pooled reactions of 28 audience members Prediction: Different Nasality > N-just > Similar Nasality (Slight VNAS + short N (Heavy VNAS + short N matched with heavier V NAS combined with Slight VNAS + short N) + long N)

Slide 21

~ Perceptual equality amongst V and N ~ Prediction: Different Nasality > N-just > Similar Nasality

Slide 22

Production and observation synopsis Summary Nasal signals in codas have worldly steadiness: settings with shorter N have more expectant vowel nasalization. The perceptual partner to (relative) transient dependability of nasalization is that audience members are touchy to aggregate nasalization over the syllable rhyme, reacting to vocalic and consonantal nasality as if perceptually equivalent.

Slide 23

The phonetic discoveries are steady with generally bore witness to phonological examples: Phonologically, nasal codas are hearty. A few dialects permit just nasals as codas, and dialects that have lost coda consonants and for which we have a definite order (e.g., Romance, Chinese; Chen & Wang, 1975 ) show that N misfortune takes after a slower direction than does oral stop misfortune. This strength is with regards to fleeting solidness of the nasal signal in

Recommended
View more...