Examining the effect of first formant transition on perception of dynamic spectral change in the second formant

2017 ◽  
Vol 142 (4) ◽  
pp. 2705-2705
Author(s):  
Amie Roten ◽  
Michelle R. Molis
2020 ◽  
Vol 63 (8) ◽  
pp. 2649-2666
Author(s):  
Jimin Lee ◽  
Heejin Kim ◽  
Yong Jung

Purpose The current study examines the pattern of misidentified vowels produced by individuals with dysarthria secondary to amyotrophic lateral sclerosis (ALS). Method Twenty-three individuals with ALS and 22 typical individuals produced 10 monophthongs in an /h/-vowel-/d/ context. One hundred thirty-five listeners completed a forced-choice vowel identification test. Misidentified vowels were examined in terms of the target vowel categories (front–back; low–mid–high) and the direction of misidentification (the directional pattern when the target vowel was misidentified, e.g., misidentification “to a lower vowel”). In addition, acoustic predictors of vowel misidentifications were tested based on log first formant (F1), log second formant, log F1 vowel inherent spectral change, log second formant vowel inherent spectral change, and vowel duration. Results First, high and mid vowels were more frequently misidentified than low vowels for all speaker groups. Second, front and back vowels were misidentified at a similar rate for both the Mild and Severe groups, whereas back vowels were more frequently misidentified than front vowels in typical individuals. Regarding the direction of vowel misidentification, vowel errors were mostly made within the same backness (front–back) category for all groups. In addition, more errors were found toward a lower vowel category than toward a higher vowel category in the Severe group, but not in the Mild group. Overall, log F1 difference was identified as a consistent acoustic predictor of the main vowel misidentification pattern. Conclusion Frequent misidentifications in the vowel height dimension and the acoustic predictor, F1, suggest that limited tongue height control is the major articulatory dysfunction in individuals with ALS. Clinical implications regarding this finding are discussed.


2003 ◽  
Vol 46 (3) ◽  
pp. 636-648 ◽  
Author(s):  
Mark S. Hedrick ◽  
Mary Sue Younger

The two aims of this study were (a) to determine the perceptual weight given formant transition and relative amplitude information for labeling fricative place of articulation perception and (b) to determine the extent of integration of relative amplitude and formant transition cues. Seven listeners with normal hearing and 7 listeners with sensorineural hearing loss participated. The listeners were asked to label the fricatives of synthetic consonant-vowel stimuli as either /s/ or /∫/. Across the stimuli, 3 cues were varied: (a) The amplitude of the spectral peak in the 2500- Hz range of the frication relative to the adjacent vowel peak amplitude in the same frequency region, (b)the frication duration, which was either 50 or 140 ms, and (c) the second formant transition onset frequency, which was varied from 1200 to 1800 Hz. An analysis of variance model was used to determine weightings for the relative amplitude and transition cues for the different frication duration conditions. A 30-ms gap of silence was inserted between the frication and vocalic portions of the stimuli, with the intent that a temporal separation of frication and transition information might affect how the cues were integrated. The weighting given transition or relative amplitude differed between the listening groups and depended on frication duration. Use of the transition cue was most affected by insertion of the silent gap. Listeners with hearing loss had smaller interaction terms for the cues than listeners with normal hearing, suggesting less integration of cues.


1987 ◽  
Vol 81 (S1) ◽  
pp. S35-S35
Author(s):  
Joan Besing ◽  
John K. Cullen ◽  
Robert J. Porter ◽  
Randal Rampp ◽  
M. Jane Collins

2020 ◽  
Vol 29 (3) ◽  
pp. 391-403
Author(s):  
Dania Rishiq ◽  
Ashley Harkrider ◽  
Cary Springer ◽  
Mark Hedrick

Purpose The main purpose of this study was to evaluate aging effects on the predominantly subcortical (brainstem) encoding of the second-formant frequency transition, an essential acoustic cue for perceiving place of articulation. Method Synthetic consonant–vowel syllables varying in second-formant onset frequency (i.e., /ba/, /da/, and /ga/ stimuli) were used to elicit speech-evoked auditory brainstem responses (speech-ABRs) in 16 young adults ( M age = 21 years) and 11 older adults ( M age = 59 years). Repeated-measures mixed-model analyses of variance were performed on the latencies and amplitudes of the speech-ABR peaks. Fixed factors were phoneme (repeated measures on three levels: /b/ vs. /d/ vs. /g/) and age (two levels: young vs. older). Results Speech-ABR differences were observed between the two groups (young vs. older adults). Specifically, older listeners showed generalized amplitude reductions for onset and major peaks. Significant Phoneme × Group interactions were not observed. Conclusions Results showed aging effects in speech-ABR amplitudes that may reflect diminished subcortical encoding of consonants in older listeners. These aging effects were not phoneme dependent as observed using the statistical methods of this study.


Languages ◽  
2021 ◽  
Vol 6 (2) ◽  
pp. 61
Author(s):  
Lisa Kornder ◽  
Ineke Mennen

The purpose of this investigation was to trace first (L1) and second language (L2) segmental speech development in the Austrian German–English late bilingual Arnold Schwarzenegger over a period of 40 years, which makes it the first study to examine a bilingual’s speech development over several decades in both their languages. To this end, acoustic measurements of voice onset time (VOT) durations of word-initial plosives (Study 1) and formant frequencies of the first and second formant of Austrian German and English monophthongs (Study 2) were conducted using speech samples collected from broadcast interviews. The results of Study 1 showed a merging of Schwarzenegger’s German and English voiceless plosives in his late productions as manifested in a significant lengthening of VOT duration in his German plosives, and a shortening of VOT duration in his English plosives, closer to L1 production norms. Similar findings were evidenced in Study 2, revealing that some of Schwarzenegger’s L1 and L2 vowel categories had moved closer together in the course of L2 immersion. These findings suggest that both a bilingual’s first and second language accent is likely to develop and reorganize over time due to dynamic interactions between the first and second language system.


2002 ◽  
Vol 698 (1-4) ◽  
pp. 243-252 ◽  
Author(s):  
T. Hatsuda

1981 ◽  
Vol 103 (1) ◽  
pp. 65-73 ◽  
Author(s):  
V. W. King ◽  
J. L. Lauer

Partial and complete emission band reversals have been observed in the infrared emission spectra from portions of operating sliding contacts. An elementary analysis has been carried out to show that partial reversals are due to temperature gradients in the fluid film—the film acts both as a radiation-emitter and absorber, and that total reversals—an emission spectrum appears as an absorption spectrum—are likely to be due to a continuum source, such as hot solid asperities. The total energy radiated under the latter conditions exceeds that under the others. A decrease in gap width with increased load was accompanied by a dramatic spectral change in the case of 5P4E polyphenyl ether, which is indicative of molecular alignment.


1997 ◽  
Vol 30 (10) ◽  
pp. 2956-2962 ◽  
Author(s):  
Kohji Tashiro ◽  
Takeru Kitashoji ◽  
Jyoji Ihata
Keyword(s):  

Sign in / Sign up

Export Citation Format

Share Document