VOT Values Research Articles

An Acoustic Study of the Stop Consonants in Lettered-Words Produced by Chinese Mandarin Speakers

This paper presents key findings from an acoustic study on Chinese Mandarin Speakers’ realization of the stop consonants in lettered-words and the stop consonants in Chinese phrases. The research was conducted through voice recording and analysis of data collected from Chinese Mandarin speakers. Recordings of 2000 overall tokens were collected and analyzed using Praat phonetic software. The mean VOT values obtained from both lettered-words and Chinese phrases were compared and statistically tested using the Independent Samples t-test. The research findings show that first, the stop consonants in lettered-words produced by Chinese Mandarin speakers are all voiceless stop consonants, distinguished as being [+aspirated] and [-aspirated]. Second, the VOT values of [p] [t] [k] in lettered-words are significantly lower than their counterparts in Chinese phrases. Third, the stop consonants in lettered-words have the same stop pattern as those in Chinese phrases. The research presents proof that where there is phonemic similarity (but phonetic dissimilarity) across Chinese Mandarin and English, L1 phonetic properties are strong for Chinese Mandarin speakers to produce lettered-words in L1 environment, which further confirms former scholars’ conviction of mother tongue interference in L2 learning. Similar phonemes in L1 and L2 are realized identically to L1 sounds, and there is significant interference from the speaker’s native language on phonetic properties produced in L2. In addition, the research presents implication for future research that might explore other acoustic features of lettered-words in Chinese Mandarin.

Acoustic correlates of oral and glottal stops in Tahitian

The Eastern Polynesian language Tahitian spoken in French Polynesia, is described as having a simple phonological inventory with a single stop series: /p, t, ʔ/. Little is known about the acoustic properties of these consonants compared to well-resourced languages. Five female speakers of Tahitian produced multiple repetitions of tokens that varied in terms of stress location and prosodic phrase position. Acoustic analyses of stop duration and VOT, as well as voicing measures including voicing fraction and strength of excitation were conducted. As predicted, oral stops were significantly longer than glottal stops by 63 ms in word medial contexts with no effect of stress. VOT values were on average 21–26 ms (with /t/ marginally longer than /p/) confirming these are short lag stops (Cho and Ladefoged, 1999). Stress was marginally significant with shorter VOT values in stressed syllable onsets with no affect of prosodic position, unlike Hawaiian (Davidson and Parker Jones, 2022). Like Hawaiian, however, (Davidson, 2021), the glottal stop in Tahitian has gradient realizations although the majority are realised with full glottal closure with low voicing fraction and strength of excitation values. There are also clear effects of greater coarticulation in unstressed versus stressed syllables suggesting localised hyperarticulation in the latter context.

A sociophoentic study of intergenerational differences in Kemie language

Previous sociophonetic studies have primarily concentrated on European languages or other dominant languages, with Chinese dialects and minority languages receiving less attention. This article aims to examine the phonetic variations of vowels, stops and tones emerged owing to age differences in Kemie Language, an Austroasiatic language in China. 18 speakers were categorized by age into young, middle-aged and older groups with 6 speakers in each group. Acoustic parameters such as formants, VOT, f0 and duration were extracted, and linear mixed effects model and repeated measures ANOVA in R were employed to test whether the above parameters were significantly different across the groups. Results show that: (1) The monophthongs /ɛ/ /a/ /ɔ/ manifest significant variations through tongue height or frontness or both. (2) VOT values of /b/ in the above three groups are significantly different and inverse with age, while no consistent conclusions can be reached in terms of /d/ /ɡ/. (3) Low-rising (13) and high-rising (35) tones with stop codas (cu yun wei) haven’t merged across the groups, but pitch range of the two tones become narrower with decreasing age; Low-rising (13) and mid-level (33) tones with sonorant codas (shu yun wei) have merged across groups.

An Acoustic Analysis of Voice Onset Time in the Production of Voiceless Plosives in English and Iraqi Kurdish

This paper is a study of voice onset time)VOT( in Iraqi Kurdish and English. The stops /t, k/ were used in initial position and were embedded in words which were presented to ten native speakers of Iraqi Kurdish and four English native speakers. Recordings were made and analyzed using a computer software that generated a spectrogram for each word (prrat). The results prove that stops in Iraqi Kurdish and in English occur at the same place along the VOT. The study also found that Place of articulation has a significant effect on VOT values and asserts the strong relationship between VOT values and place of articulation. Moreover vowel context has its significant effect on VOT values and asserts the strong relationship between VOT values and vowel context. Additionally, it is found that gender impact on VOT values varied from one language group to another, gender has no significant effect in the production of voiceless plosives in Iraqi Kurdish. Whereas English male speakers produce longer voice onset time than Female speakers.

An Acoustic Analysis of Voice Onset Time in the Production of Voiceless Plosives in English and Iraqi Turkmen

This paper is a study of voice onset time, VOT, in Iraqi Turkmen and English. The stops /t, k/ were used in initial position and were embedded in words which were presented to ten native speakers of Iraqi Turkmen and four English native speakers. Recordings were made and analyzed using computer software that generated a spectrogram for each word (prrat). Results prove that stops in Iraqi Turkmen and in English occur at the same place along the VOT. The study also found that Place of articulation has a significant effect on VOT values and asserts the strong relationship between VOT values and place of articulation. Moreover vowel context has its significant effect on VOT values and asserts the strong relationship between VOT values and vowel context. Additionally, it is found that gender impact on VOT values varied from one language group to another, gender has no significant effect in the production of voiceless plosives in Iraqi Turkmen. Whereas English male speakers produce longer voice onset time than Female speakers.

Voice onset time of plosive consonants in initial position in badidni dialect

Phoneticians have presented several ways and theories for knowing the features and characteristics of plosive consonants. Acoustic ways are ones that languages’ plosive consonants are determined by them. In the last 60 years VOT was widely used as an acoustic parameter by phoneticians for comparing voicing contrast and determining the categories and features of plosive consonants as an acceptable way. Until now it’s not known that plosive consonants are determined in badini dialect according to the VOT parameter. This study is an attempt to what extent we can benefit from the mentioned parameter for specifying voicing contrast distinction, voicing categories and their properties and features. For this purpose the two plosive consonants /b, p/ are used as a stimulus of this investigation in the initial 6 monosyllabic words in the environment of the two vowels /ɑ æ,/. The collected data are taken from (20) badini dialect speakers. It’s assumed that voicing contrast, VOT values and the categories of plosive (voiced and voiceless) consonants are different in baddini dialect. Depending on procedures and statistical analysis tests it concludes that; VOT is formed in three ways and three separate values. This study consists of five sections; the first section is the introduction, the second one is theoretical framework, in the third section the procedures of doing this study are shown, statistical results and discussing them are mentioned in the fourth section and the conclusions of the study are shown in the fifth section.

Speech perception and children with speech sound disorder: An assessment of non-errored speech sounds

Previous studies assessing speech perception in children with speech sound disorder (SSD) suggest inconsistent, if any, differences from typically-developing peers (TD). However, others suggest that children with SSD perceive inaccurate productions as acceptable variants of a sound. Currently, limited research explores speech perception of non-errored sounds in children with SSD compared to TD peers. This study assesses how distributional characteristics of naturally produced child speech stimuli, collected from six 2–3-year-old English-speaking children, might impact listening behavior of children with SSD for non-errored sounds. Stimuli included six exemplars per speaker and VOT categories of short-lag /b d/, short-lag /p t/, long-lag /b d/, long-lag /p t/. For each POA and VOT group, /b d/ and /p t/ VOTs were bimodally distributed (shorter for voiced targets), separated by a minimum 5 ms gap. We will compare perception results for TD children and children with SSD (ages 6;0–10;11). Earlier findings showed high accuracy in adults and TD children for targets with appropriate VOT values with adults slightly outperforming children. We hypothesize that children with SSD will demonstrate similar accuracy levels between groups when VOT values are appropriate for the target, and slightly greater response variability and group differences for inappropriate VOTs.

Investigating the nature of speech representations by contrasting computational models against behavioral data

How do listeners interpret speech input when exposure does not contain information about the talker’s category representations? Competing accounts are rarely contrasted. We implement competing hypotheses within the same general computational framework (Bayesian inference). All models were trained on phonetic productions to predict perception, reflecting the hypothesis that listeners learn category representations from the speech input. We compare them against two experiments on the perception of L1-US English stop voicing (N = 24 and 122). Both experiments used minimal pairs (e.g., tin/din) varying in the primary cue (VOT) while keeping secondary cues (f0 and vowel duration) at expected correlations. VOT values occurred equally often and spanned the range observed in US English. We find that (1) models that integrated perceptual noise performed better in predicting categorization responses than those without; (2) models with multiple cues performed better than those with just VOT; (3) models trained on talker-normalized phonetic cues performed better than those trained on unnormalized cues; and, surprisingly, (4) models that also normalized the novel speech input during the experiment performed worse than those that did not. (3) and (4) suggest that listeners’ long-term representations are based on talker-normalized cues but require *labelled* input—contrary to most normalization accounts.

Young Pashtun L1 speakers' laryngeal contrast and voicing variation in Korean word-initial stops production

While age proved to have a critical influence on learning L2 phonology, the reason behind it is yet controversial. (brain maturation, LOR, quality l2 input, crib or sequential bilingualism, etc.) This paper examines 6 young people in the same household using Pashtun as their L1. As the elder children arrived in Korea at ages 10, 8, 6, and 3 respectively, the younger two were born in Korea now aged 10 and 8. All family members have equal LOR(10 years except for the youngest), equal L1, and equal L1 and L2 use extent in the home. Examining the differences between VOT and f0 production in these young people will clarify some of the explanations behind the age effects. Moreover, the data reveals the unique L1 transfer phenomena. While Pashtun is a true voicing language, Korean is an aspirating language where all 3- way contrast stops have positive VOT values. This can lead the young people to produce word-initial Korean stops in intersonorant positions with a marked voicing-ratio and to produce allophones that neither exists in their L1 nor in L2 such as voiced aspirated stop sounds ( bɦ, d̪ɦ, gɦ). This paper will also examine whether these young people’s VOT and f0 productions would align with the current tonogenesis like sound change in Korean. This paper used picture naming tasks with easy words instead of reading non-words to not test their grapheme-phoneme correspondences. Each person produces 9 phonemes of word-initial Korean stops (3-way laryngeal contrasts in 3 POA) both in the position in isolation and carrier sentence.

Arabic emphatic consonants as produced by English speakers: An acoustic study

This study examines the production of emphatic consonants as produced by American L2 learners of Arabic. To this end, 19 participants, 5 native speakers and 14 L2 learners, participated in a production experiment in which they produced monosyllabic CVC pairs that were contrasted in terms of whether the initial consonant was plain or emphatic. The acoustic parameters that were investigated are VOT of voiceless stops, COG of fricatives, and the first three formant frequencies of the target vowels. The results of the native speakers showed that VOT is a reliable acoustic correlate of emphasis in MSA. The results also showed that vowels in the emphatic context have higher F1 and F3 and lower F2. The results showed that the L2 learners produced comparable VOT values to those of native Arabic speakers. Further, L2 learners produced a significantly lower F2 of the vowels in the emphatic context than that in the plain context. Proficiency in Arabic played a role on the F2 measure; the intermediate learners tended to be more native-like than the beginning learners. As for F3, the results of the L2 learners unexpectedly showed that the beginning learners produced a higher F3 in the context of fricatives only. This suggests that the relationship between emphasis and proficiency depends on whether the preceding consonant is a stop or fricative.

Gender effect on the use of phonetic cues for stops in the Kyungsang dialect

The present study aimed to examine whether there was a gender effect on the phonetic properties of Korean stops in the Kyungsang dialect and investigate how the genders used the acoustic cues across different prosodic positions and focus conditions. Voice onset time, the peak of fundamental frequency, and vowel duration were compared in initial and medial positions of IP and under focused and unfocused conditions. In general, the VOT values of lenis and aspirated stops were shown to be longer for Kyungsang male speakers than for female speakers. The vowel duration, F0 and semitones of the two stops were greater for the female speakers than for the male speakers, and the systematic difference of semitones depending on different prosodic conditions was only found in the speech of the female speakers. While the Kyungsang male speakers made use of VOT to distinguish stop categories, the Kyungsang female speakers put more weight on the use of F0 to contrast the stops across the different prosodic conditions.

Destabilizing effects of L2 explicit pronunciation instruction on L1 speech:

Anchored in a Complex Dynamic perspective on language development, this study explores the effects of L2 explicit pronunciation instruction on L2 English and L1 Brazilian Portuguese VOT production. To this end, 16 Brazilian intermediate users of L2 English were investigated. Volunteers were divided into control and experimental groups. The latter received explicit pronunciation instruction on the production of the English voiceless stops. The study included three data collections, a pre-test, an immediate post-test and a delayed post-test. An acoustic analysis of VOT duration is reported. The results revealed that the control group did not produce the expected VOT pattern for L2 English at any time of the study. No alterations in the L1 were reported. On the other hand, after the instruction, the experimental group produced higher L2 English VOT values. First language attrition was reported since the L1 BP VOT durations also increased after instruction. In addition to highlighting the effects of explicit instruction on L2 development and L1 attrition, this study confirms that language development is constant, and that even L1 adult grammars are not rigid, with potential to change due to the perception and processing of novel nonnative phonetic-phonological categories.

Examining the link between speech perception and production through categorical boundaries

The link between speech perception and production is one of the central topics in speech perception theories. Newman (2003) examined the correlations between acoustic measures of listeners’ perceptual and production prototypes for a given speech category, and found significant correlations for stop consonants (VOT) and for voiceless fricatives (spectral peaks). Shultz et al. (2012) also investigated the possible link between perception and production by examining speakers’ relative cue weighting of VOT and f0 in word initial stops. Although their results showed no significant correlation between weighting of VOT in production and perception, they reported a trend toward a slightly positive correlation between production and perception for VOT values. The current study extends these findings and aims to explore the perception-production link by examining the relationship between the location of categorical boundaries (between /p/ and /b/) in perception and production. If mental representations for perception and production are directly linked, then speakers with higher VOT values in both categories in production would be expected to also show a higher categorical boundary in perception.

Phonological interaction in Spanish-English bilinguals: effects of cognate usage on voice onset time of voiced stops

ABSTRACT This study compared Spanish-English bilinguals’ and English monolinguals’ VOT values for /b, d, g/ in cognates and non-cognates. Twenty-six young adult participants (fourteen bilinguals and twelve monolinguals, mean age 24 years) were administered a picture-naming task, balanced for cognate and non-cognate forms. VOT values of 30 target words per participant and per language were measured. The data support cross-linguistic interaction in bilinguals’ phonologies and indicate that cognates can interfere with bilinguals’ phonemic distinctions. In English, bilinguals’ VOTs exhibited significantly greater lead voicing than monolinguals’, and, for all participants, /b/ had longer lead voicing than /d/ and /g/. Comparing bilinguals’ VOTs for Spanish versus English revealed significant differences by language (English shorter), between /b/ and /g/, and between cognates and non-cognates, with shorter lead voicing in cognates than in non-cognates. More detailed results showed that in bilinguals’ Spanish productions with word-initial /d/ and English productions with word-initial /b/, cognates exhibited shorter lead voicing (more English-like) than non-cognates. The conclusion is that the bilinguals’ VOTs exhibited some cross-linguistic influence in relation to cognate usage, in the direction toward their dominant language, English. Further analyses also revealed a role of current daily usage in VOT performance.

Effects of F0 and VOT on the Identification of Korean Word-Initial Oral and Nasal Stops

The paper examines which acoustic cues affect Korean"s identification of word-initial oral and nasal stops (/p, ph, p", t, th, t", m, n/) by manipulating stops in terms of F0, VOT, and nasal consonantal portion. F0 values were critical for the identification of lax, aspirated, and nasal stops and the change of their F0 values resulted in more than 97% misperception of lax to aspirated stops, aspirated to lax stops, and nasal to tense stops. VOT values were critical for the identification of tense stops: the tense stimulus with lengthened VOT have never been perceived as tense. Especially, in the identification of nasals, the nasal stimulus lacking any nasal consonantal portion were identified as nasals to 88% of Korean listeners, and the stimulus manipulated by high F0 was perceived as 100% of tense stops. The results indicate that F0 is a major acoustic cue for the contrast among word-initial nasals and oral stops, and that low F0 after nasals rather than nasal portions is a more important cue for the identification of word-initial nasals.

Phonological memory traces do not contain phonetic information

We use a "varying standards" oddball paradigm and compare two phonetically differing conditions to find evidence that the auditory cortex has access to discrete phonological representations when making predictions about incoming speech sounds. Brain responses were recorded with a 128-electrode EEG system as subjects passively listened to synthetic speech sounds from a /dæ/-/tæ/ continuum. Deviant stimuli were compared to a control condition to obtain a mismatch negativity (MMN) response, indicative of a "surprise" at a stimulus that deviates from the memory trace. We tested two conditions with phonetically different varying standards - a "low-T" condition with VOT values of 60, 65, and 70 ms, and a "high-T" condition with VOT values of 75, 80, and 85 ms. The amplitude of the mismatch response is generally sensitive to acoustic distance. If the memory trace generated by the auditory cortex contains phonetic information about the standards, then a difference in mismatch amplitude is expected. However, if a phonological memory trace is used, every standard will be represented identically and no difference in mismatch amplitude will be observed. We found no distance effect in our results, indicating an absence of fine-grained phonetic information. This supports the claim that the auditory cortex has access to phonological category representations.

한국인 학습자의 일본어 어두 유성파열음의 발음 습득 실태 : 자발(회화) 음성의 VOT를 중심으로

This study describes the acquisition status of Japanese initial voiced plosive of Korean learners using spontaneous speech, the descriptive contents and the pronunciation teaching method of Japanese initial voiced plosive’s VOT in commercial Japanese voice textbooks. As a result, the following findings were confirmed.BR First, most of Korean learners were acquiring Japanese initial voiced plosive as +VOT. The longest +VOT value appears in the phonetic environment where the point of articulation is the soft palate and the subsequent vowel is /i/ whereas the longest -VOT value appears when the point of articulation is the alveolar ridge and the subsequent vowel is /e/.BR Second, the VOT value of Japanese initial voiced plosive by both Korean learners and Japanese speakers is most distributed in common in the range of 0ms to 50ms , but in terms of the proportion , Korean learners’ VOT values were higher than Japanese speakers`.BR Third, commercial Japanese voice textbooks describe Japanese initial voiced plosive as -VOT, and suggest learners pronounce it by adding ‘ん’ or a vowel before the voiced stop. But, it is necessary to modify and complement the explanation and the pronunciation teaching method of VOT considering the fact that +VOT appears in the pronunciation of most Japanese speakers and Korean learners, and it is also necessary to examine the validity and the need for improvement of existing pronunciation teaching methods.

Effects of L1 orthography and L1 phonology on L2 English pronunciation

Des recherches récentes ont dévoilé les effets de l’orthographe sur la durée consonantique en anglais L2 de locuteurs natifs italiens (le [p] étant réalisé plus long pour floppy que pour copy ). Nous avons comparé cet effet orthographique avec un effet de la phonologie de la L1 indépendant de l’orthographe, le Voice Onset Time (VOT). La tenue et le VOT d’occlusives dans des mots-cibles ont été mesurés chez 30 apprenants d’anglais L2 en Italie, 30 bilingues tardifs italiens résidant au Royaume Uni et 30 locuteurs natifs de l’anglais. Si les valeurs de VOT produites par les bilingues se différencient de manière significative par rapport à celles des apprenants, les valeurs de tenue restent comparables dans les deux groupes. En outre, le taux d’adaptation du VOT en anglais L2 est plus important chez les bilingues tardifs que chez les apprenants. Cela prouve que les effets orthographiques sur la longueur consonantique peuvent être plus résistants à l’exposition de la L2 que les effets de la phonologie de la L1 indépendants de l’orthographe.

VOT of Stops by Mandarin-Japanese Bilingual Children

This study investigated VOT of word-initial Japanese and Mandarin stops produced by Mandarin-Japanese bilingual children in order to observe bilinguals' phonetic systems. Examination of Japanese and Mandarin stop productions at the same articulation points in bilingual children aged three to six years old revealed important findings: (i) They tended to lengthen VOT values at the age of four and five in Japanese stop production, and at the age of five in Mandarin; and (ii) They maintained phonetic contrast in three VOT categories in their two languages: a) voiced stops in Japanese/unaspirated stops in Mandarin, b) voiceless stops in Japanese, and c) aspirated stops in Mandarin. These VOT patterns lend credibility to the idea that early bilingual children develop distinct Japanese and Mandarin phonetic systems, while showing cross-linguistic influence in each language. The findings of this study could provide a better understanding of the phonological representation in early bilinguals.

VOT in Tibetan is conditioned by tone

Several South Asian languages exhibit interactions among tone and consonant phonation, including Tamang (Mazoudon 2014) and Gurung (Ronkos 2015). In Central Tibetan, oral stops present a four-way contrast between two voicing/aspiration categories split across two lexical tones, high and low/rising. Across tones, previous analyses propose either a two-way contrast between unaspirated and aspirated stops (e.g., Dawson 1980), or a three-way contrast between voiced, unaspirated, and aspirated series (e.g., Denwood 1999), though acoustic measurements are lacking. This study recorded 19 native speakers of Tibetan in Kathmandu, Nepal, producing words exemplifying the phonological contrasts in stops and high/low tones. VOT measurements showed similar values for the unaspirated stops in both tonal environments, but a longer VOT in high-tone aspirated stops (46 ms) than low-tone aspirated stops (30 ms), or three different positive VOT values. That this three-way surface contrast cross-cuts tonal categories suggests an underlying two-way aspiration contrast. The apparent contrast between middle-VOT and long-VOT series results from the presence of a phonologically-specified high tone, which causes a lengthened VOT. This provides evidence for the conditioning of apparently-segmental characteristics by tone.Several South Asian languages exhibit interactions among tone and consonant phonation, including Tamang (Mazoudon 2014) and Gurung (Ronkos 2015). In Central Tibetan, oral stops present a four-way contrast between two voicing/aspiration categories split across two lexical tones, high and low/rising. Across tones, previous analyses propose either a two-way contrast between unaspirated and aspirated stops (e.g., Dawson 1980), or a three-way contrast between voiced, unaspirated, and aspirated series (e.g., Denwood 1999), though acoustic measurements are lacking. This study recorded 19 native speakers of Tibetan in Kathmandu, Nepal, producing words exemplifying the phonological contrasts in stops and high/low tones. VOT measurements showed similar values for the unaspirated stops in both tonal environments, but a longer VOT in high-tone aspirated stops (46 ms) than low-tone aspirated stops (30 ms), or three different positive VOT values. That this three-way surface contrast cross-cuts tonal categories suggests...