Vowel Duration Ratios Research Articles

Articulatory Characteristics as Functions of Articulation Distance and Articulatory Direction in Children with Cochlear Implants

Objectives: Although children with cochlear implantation use correct pronunciation, many of them have difficulties with producing details of speech. The main aim of this study was to measure the articulation features of speech in well-performing children with cochlear implants (CIs).Methods: The participants were 15 children with CIs, with a mean age of 7 years (range: 5-10 years). A control group of children with typical hearing (TH) was included. Nonwords were manipulated in the form of different articulation distances and articulation movement directions. We measured phoneme accuracy, the ratio of vowel duration between the first and the second consonant, and the slope of the second formant transition (F2).Results: The two groups showed similar performance in the phoneme accuracy and the ratio of vowel duration, but the CI group produced sensitive changes according to the first consonant of nonwords and the articulation distance of the consonant pair. The CI group showed a higher slope of F2 than the TH group.Conclusion: These results suggest that children with CIs can control a range of motion and the speed of articulation organs, but their articulatory control abilities are less sophisticated and less stable. This study is meaningful in that it examined the articulation manipulation ability of children with cochlear implantation, which is perceptually difficult to distinguish, through an acoustic-phonetic evaluation method.

What Makes Spanish-Accented English Sound Spanish-Accented? Acoustic Measures and Listener Cues

This study aimed to understand which acoustic parameters of Spanish-accented English are correlated with listeners’ perception of Spanish-accentedness. Temporal differences were analyzed in multisyllabic target words spoken in sentences by 22 Spanish speakers of English and five native speakers of American English (AE). Recordings were presented to AE listeners who judged the degree of accentedness on a 9-point scale. Spearman rank order correlation showed that the listeners’ ratings of degree of accentedness in sentences correlated strongly (r= +0.82) with those in words. Listeners’ ratings of accentedness correlated in varying degrees with various temporal measures, namely Overall word durations (+0.04 to +0.56), Stressed/unstressed vowel duration ratios (–0.01 to +0.35), Voice Onset Time of stops (+0.26 to +0.36), and, closure duration (+0.29 to +0.59). Results suggest that Spanish-accented English is characterized by systematic temporal differences from native AE, and that these temporal differences contribute to the perception of accentedness. Implications of findings in improving theoretical understanding and applied practices are discussed.

The effect of word-final stops’ voicing on the vowel duration and its relation with Korean speakers' english proficiency

It has been well known that the vowels preceding voiced stops are produced durationally longer than the vowels preceding voiceless stops in English. Based on this phonetic phenomenon, this research investigates to determine whether there is any relation between the Koreans' English proficiency and the difference in the ratio of the English vowel duration before the voiced stops to the English vowel duration before the voiceless stops. For analysis, the Korean-Spoken English Corpus (K-SEC, made in 2004) and the Rated Korean-Spoken English Corpus (Rated K-SEC, made in 2017) were utilized where the 32 Korean elementary school students from Seoul and Gyeonggi province articulated the target items such as bag/back and cab/cap. These speakers were categorized into one of the three different proficiency groups (Advanced, Intermediate, and Novice) according to their speaking ability. From the findings, native English speakers were found producing the vowels before the word-final voiced stops 1.421 times longer than the vowels before the word-final voiceless stops. Similarly, in the case of the Korean speakers, they produced the vowels before the word-final voiced stops longer than the vowels before the word-final voiceless stops. However, the study couldn't discover any possible relationship between their English proficiency and the ratio of the vowel duration before the word-final voiced stops to the vowel duration before the word-final voiceless stops in English (the ratio—Advanced: 1.275, Intermediate: 1.209 and Novice: 1.25).

Production of pre-fortis clipping in English by Korean and English speakers

The term pre-fortis clipping refers to the phenomenon in English of vowels being shorter before voiceless obstruents than before voiced ones. This paper examines whether ratios of duration for vowels before voiceless versus voiced obstruents are affected by speakers’ L1 and segmental properties such as consonantal manner of articulation, vowel height, and phonemic vowel length. Nine Korean speakers and four native speakers of English were recorded reading 30 minimal pairs consisting of English monosyllabic CVC words differing in coda voicing. The mean ratio of duration for vowels preceding voiceless and voiced obstruents respectively was greater for Korean speakers than for English speakers, but mixed-effects model analyses with speaker and item random effects indicate that the effect of L1 was not significant, though there was greater variation for Korean speakers than for English speakers. The ratio of vowel duration significantly differs between Korean and English speakers only before fricatives. Vowels before voiced fricatives are not sufficiently long for Korean speakers as compared to English speakers. It is suggested that, due to their lack of experience with released fricatives, Korean speakers fail to vary vowel durations sufficiently depending on the consonant manner, before stops and fricatives.

Acoustic quantification of /i/-/I/ overlap in children 21 to 33 months

This longitudinal investigation examined the temporal and spectral characteristics of the high front vowels /i/ and /I/ as produced by nine monolingual US English children from 21–33 months. Vowel overlap was quantified in two-dimensional (F1, F2) and three-dimensional (F1, F2, duration) space using Spectral Overlap Assessment Measure (SOAM). These findings were compared with the results from Support Vector Machine (SVM) vowel classification, vowel duration ratios, and measures of effect size, to determine whether a spectral/temporal trading effect existed in the early vowel productions of young children. Children between the ages of 21 and 33 months are highly variable in the way they use spectral and temporal parameters to distinguish between these two adjacent vowels. However, findings pointed to the existence of a spectral/temporal trading effect when spectral overlap values are relatively high (>60%) at 21 and 24 months of age.

Temporal features of Spanish-accented English

In this study we attempt to identify the acoustic parameters of Spanish accented English that give it the perception of being accented. Recordings of eight multisyllabic (3, 4, and 5 syllables) words spoken in sentences by 22 Spanish speakers of English and five native speakers of American English (AE) were analyzed for temporal acoustic differences. Segmental durations of Spanish productions, including differences of stressed–unstressed (s/u) vowel duration ratios, Voice Onset Time (VOT) and flap/stop duration, differed from native productions. The sentences and the excised target words were given to native (AE) listeners (N=10) who judged the degree of accentedness on a 9-point scale. The degree of accentedness on target words correlated with perceived ratings on the eight sentences (i.e., global accentedness), rho=+0.82. Spearman rank-order correlations between s/u vowel duration ratios and native listener ratings of accentedness on target words varied from +0.04 to 0.68. The VOT duration of initial voiceless stops correlated positively with accentedness ratings for three of the four words (+0.26 to +0.35). Overall, results suggest that Spanish-accented English is characterized by significant temporal differences from native American English, and that these temporal differences contribute to the perception of accentedness as judged by native American English listeners.

Relationships between acoustic and perceptual measures of stress in nonwords produced by children with developmental apraxia of speech and children with phonological disorders

This study investigated relationships between acoustic measures of stress in nonwords produced by children with suspected developmental apraxia of speech (sDAS) and children with phonological disorders (PD), and perceptual judgments made by trained listeners. Five children with sDAS and five children with PD produced multiple tokens of eight disyllabic nonwords with iambic and trochaic stress patterns. Measures of vowel duration, f0, and intensity were made. Children from both groups produced acoustic differences between stressed and stressless syllables. In a perception experiment, 10 listeners judged whether tokens were produced with stress on the initial or final syllable. Listeners judged that children with sDAS accurately matched target stress contours less often than children with PD. The proportion of listeners judging that a word was produced with final syllable stress was used as the dependant variable in a regression, in which the ratio of vowel duration, f0, and intensity between the two syllables served as the independent variables. All three variables predicted a significant proportion of variance in listener judgments, although the overall proportion of variance accounted for was low (R2=0.212). Thus, children with sDAS are able to produce contrasts between stressed and stressless syllables, but these are often imperceptible to listeners.

Acoustic evidence for gestural overlap in consonant sequences

Acoustic evidence for temporal overlap of the two closure gestures in the environment VC ≠ CV was investigated. It was hypothesized that evidence of C 2 would be found in the VC formant transitions and would increasingly dominate the transitions as rate (and, by hypothesis, overlap) increased. Twenty repetitions (10 at a normal rate and 10 at a rapid rate) of word pairs where the first word ended in /d/ and the second began with /p/, /t/, or /k/ were elicited in a sentence context from four subjects. F 2 and F 3 transitions from the midpoint of V 1 , to just before closure were then measured. In all environments, C 2 had a clear influence on the VC formant transitions. The rate effects were less clear. For the /d ≠ k/ environment, a significant correlation was found between more prominent velar transitions and increasing ratio of vowel duration to consonant closure duration, which may be considered a measure of increasing consonant overlap. The acoustic influence of C 2 on V 1 , suggests considerable temporal overlap of the two closure gestures, and at least for the /d ≠ k/ case, increasing overlap as a function of one strategy for increasing rate.

Voicing contrasts and their phonological implications in the speech of articulation-disordered children.

Five-year-old articulation-disordered children, some classified as substituters and some as syllable reducers, were compared with normal child and adult controls in their production of voicing contrasts. These contrasts occurred in minimal pairs containing word-final obstruents and in minimal triples containing word-initial stops and /s/-plus-stop clusters in initial position. Measures of vowel duration, voice onset time (VOT), and frequency of use of phonetic voicing were made from spectrograms. In every comparison the substituters' performance resembled that of the normal controls, as did the syllable reducers' use of VOT in stop singles. The syllable reducers used larger vowel duration ratios than the normal controls in a few minimal pairs and used phonetic voicing less often in word-initial /b d g/. The production data and previously reported perception data were examined for evidence that individual syllable reducers had voicing contrasts in underlying phonological form despite their deletions of obstruents in which these contrasts occurred. Most of the syllable reducers appeared to recognize underlying voicing contrasts in at least a few final obstruents and in some of the initial stop singles.

Voicing-dependent vowel duration in English and French: Monolingual and bilingual production

It is well established in English phonology that vowels preceding voiceless consonants are approximately two-thirds the duration of vowels preceding voiced consonants when these vowels occur in stressed prepausal environments. It is claimed that such a large voicing-dependent effect does not exist in certain other languages, such as French. However, in previous studies, the syllabic and phonemic structures of the stimuli used for cross-language comparison have not been held constant. Therefore, it was the purpose of this study to compare durational effects in English and French stimulus words having the same syllabic structure and, to the extent possible, the same phonemic structure. An additional issue of the study was the ability of bilinguals to utilize the English durational rule. Computer analysis of 854 stimulus words revealed that (1) there were larger context-dependent differences in preconsonantal vowel duration in English than in French, and (2) the English and French vowel duration ratios of French–English bilinguals were essentially like those of French monolinguals. However, the bilinguals’ English vowels were longer than their French vowels.