Journal of Phonetics最新文献_第2页

Cross-linguistic similarity in L2 suprasegmental learning: evidence from Chinese learners’ perception of Japanese pitch accents 二语超片段学习中的跨语言相似性：来自中国学习者对日语音高口音感知的证据

IF 2.4 1区文学 0 LANGUAGE & LINGUISTICS

Journal of Phonetics

Pub Date : 2025-11-01 DOI: 10.1016/j.wocn.2025.101458

Yu Yang , Longjie Dong , Quansheng Xia , Yuxiao Yang , Fei Chen

The acquisition of suprasegmental features in a second language (L2), like lexical tone and pitch accent, can be challenging, yet the impact of cross-linguistic similarity on learning these suprasegmental features has been underexplored. This study explored the role of cross-linguistic similarity in Chinese learners’ perception of Japanese pitch accents, aiming to verify the Perceptual Assimilation Model for Suprasegmentals (PAM-S). In experiment 1, 25 Chinese learners of Japanese with lower proficiency level and 24 learners with higher proficiency level completed a perceptual assimilation task (PAT) that examined the cross-linguistic perceptual similarity between Mandarin tones and Japanese pitch accents. In experiment 2, the same Chinese groups and 35 native Japanese listeners completed a perceptual discrimination test (PDT) of Japanese pitch accent contrasts. Results of PAT showed that Chinese learners successfully categorized Japanese pitch accents into their native Mandarin tone categories: they perceived Japanese H*L as Mandarin Tone 4 (falling tone), LH* as Tone 2 (rising tone), and LH as Tone 1 (level tone). Moreover, results of PDT showed that Chinese learners were able to discriminate H*L–LH* and H*L–LH but had difficulty in the discrimination of LH*–LH. The results also show that Chinese learners’ ability to discriminate Japanese pitch accent contrasts did not improve consistently with increased Japanese experience. This study argues that the LH*–LH contrast is hard for L2 learners regardless of their L2 experience, because of these two accents’ acoustic similarity. The results extended the PAM-S, suggesting that L2 speech perception could be influenced by both the L1–L2 assimilation patterns and acoustic similarity.

在第二语言（L2）中习得超分段特征，如词汇语调和音高口音，可能是具有挑战性的，然而跨语言相似性对学习这些超分段特征的影响尚未得到充分探讨。本研究探讨了跨语言相似性在中国学习者对日语音高口音感知中的作用，旨在验证超音段感知同化模型（PAM-S）。在实验1中，25名日语水平较低的中国学习者和24名日语水平较高的学习者完成了一项感知同化任务（PAT），以检验普通话声调和日语音高口音之间的跨语言感知相似性。在实验2中，同样的中国小组和35名日本本土听众完成了日语音高口音对比的感知辨别测试（PDT）。PAT结果表明，中国学习者成功地将日语音高划分为母语普通话声调类别：他们将日语的H*L理解为普通话声调4（降调），将LH*理解为普通话声调2（升调），将LH理解为普通话声调1（平调）。此外，PDT结果显示，中国学习者能够区分H* L-LH *和H* L-LH，但在区分LH* -LH方面存在困难。结果还表明，中国学习者区分日语音高口音对比的能力并没有随着日语经验的增加而持续提高。本研究认为，LH* -LH的对比对于L2学习者来说很难，无论他们的L2经验如何，因为这两种口音在声学上相似。结果进一步扩展了PAM-S，表明L2语音感知可能受到L1-L2同化模式和声学相似性的影响。

{"title":"Cross-linguistic similarity in L2 suprasegmental learning: evidence from Chinese learners’ perception of Japanese pitch accents","authors":"Yu Yang , Longjie Dong , Quansheng Xia , Yuxiao Yang , Fei Chen","doi":"10.1016/j.wocn.2025.101458","DOIUrl":"10.1016/j.wocn.2025.101458","url":null,"abstract":"<div><div>The acquisition of suprasegmental features in a second language (L2), like lexical tone and pitch accent, can be challenging, yet the impact of cross-linguistic similarity on learning these suprasegmental features has been underexplored. This study explored the role of cross-linguistic similarity in Chinese learners’ perception of Japanese pitch accents, aiming to verify the Perceptual Assimilation Model for Suprasegmentals (PAM-S). In experiment 1, 25 Chinese learners of Japanese with lower proficiency level and 24 learners with higher proficiency level completed a perceptual assimilation task (PAT) that examined the cross-linguistic perceptual similarity between Mandarin tones and Japanese pitch accents. In experiment 2, the same Chinese groups and 35 native Japanese listeners completed a perceptual discrimination test (PDT) of Japanese pitch accent contrasts. Results of PAT showed that Chinese learners successfully categorized Japanese pitch accents into their native Mandarin tone categories: they perceived Japanese H*L as Mandarin Tone 4 (falling tone), LH* as Tone 2 (rising tone), and LH as Tone 1 (level tone). Moreover, results of PDT showed that Chinese learners were able to discriminate H*L–LH* and H*L–LH but had difficulty in the discrimination of LH*–LH. The results also show that Chinese learners’ ability to discriminate Japanese pitch accent contrasts did not improve consistently with increased Japanese experience. This study argues that the LH*–LH contrast is hard for L2 learners regardless of their L2 experience, because of these two accents’ acoustic similarity. The results extended the PAM-S, suggesting that L2 speech perception could be influenced by both the L1–L2 assimilation patterns and acoustic similarity.</div></div>","PeriodicalId":51397,"journal":{"name":"Journal of Phonetics","volume":"113 ","pages":"Article 101458"},"PeriodicalIF":2.4,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145424455","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"文学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Gestural restructuring beyond coarticulation in Korean /w/-vowel sequences: Evidence from phonetic, dialectal, and gender variation 韩语/w/-元音序列的手势重组超越协同发音：来自语音、方言和性别差异的证据

IF 2.4 1区文学 0 LANGUAGE & LINGUISTICS

Journal of Phonetics

Pub Date : 2025-11-01 DOI: 10.1016/j.wocn.2025.101456

Dae-yong Lee , Sahyang Kim , Taehong Cho

This study examines the articulatory patterns of Korean /w/-vowel sequences by comparing tongue dorsum movement trajectories with those of corresponding plain vowels, using Electromagnetic Articulography data from 48 speakers of Seoul and North Gyeongsang dialects. The central question is whether these sequences reflect mere coarticulation or exhibit signs of gestural restructuring in the nucleus vowel. Results reveal gradient restructuring shaped by vowel constriction degree, dialect, and gender. High vowels (/wi/-/i/) show minimal divergence, mid vowels (/we/-/e/, /wɛ/-/ɛ/) moderate divergence, and low back vowels (/wa/-/a/, /wʌ/-/ʌ/) the greatest divergence—especially in dialect- and gender-specific ways. Further analysis of the /e/-/ɛ/ merger and the recent /ʌ/-/ɨ/ split in North Gyeongsang sheds light on how vowel distinctions interact with /w/. The /we/-/wɛ/ pair shows a stronger merger than /e/-/ɛ/, supporting the view that /w/ triggers gestural restructuring of the nucleus vowel and thus plays an active role in reshaping merger trajectories. This effect is further illustrated by the /wa/-/wʌ/ and /a/-/ʌ/ contrasts, with a stronger merger in the /w/-initial context—an effect notably led by male speakers. Interestingly, North Gyeongsang males preserve the /a/-/ʌ/ contrast more robustly than the /wa/-/wʌ/ contrast, possibly due to hyperarticulation of a phonetically redefined /ʌ/ resulting from the recent /ʌ/-/ɨ/ split. These findings are interpreted within a dynamical framework of gestural blending strength (GBS), which varies by vowel constriction and coarticulatory resistance but remains stable for /w/. Overall, the results suggest that what may have begun as low-level coarticulation has evolved into systematic gestural restructuring—a gradient shift toward phonological reorganization shaped by phonetic context, sound change, and sociophonetic variation.

本研究利用来自首尔和庆尚北道方言的48名说话者的电磁发音数据，通过比较舌背运动轨迹和相应的普通元音，研究了韩语/w/-元音序列的发音模式。核心问题是，这些序列是否仅仅反映了协同发音，还是在核心元音中表现出手势重组的迹象。结果表明，元音收缩程度、方言和性别决定了梯度重构。高元音（/wi/-/i/）分化最小，中元音（/we/-/e/, /w / /-/e/）分化适中，而低后元音（/wa/-/a/, /w / /-/ w / /）分化最大——尤其是在方言中——并且以性别为特征。进一步分析庆尚北道的/e/-/ j /合并和最近的/ j / /-/ j /分裂，可以揭示元音差异是如何与/w/相互作用的。与/e/-/ w/相比，/we/-/w /对的合并更强，这支持了/w/触发核心元音的手势重组，从而在重塑合并轨迹中发挥积极作用的观点。这种效果进一步体现在/wa/-/w /和/a/-/ w/的对比中，在/w/-开头的语境中合并更强，这种效果明显由男性说话者主导。有趣的是，庆尚北道的男性保留/a/-/ / /的对比比/wa/-/w /的对比更强烈，这可能是由于最近的/ / /-/ / /分裂导致的语音上重新定义的/ /的高清晰度。这些发现是在手势混合强度（GBS）的动态框架内解释的，GBS随元音收缩和协同发音阻力而变化，但对于/w/保持稳定。总的来说，研究结果表明，可能从低水平的协同发音开始，已经演变成系统的手势重组——一个由语音语境、声音变化和社会语音变化形成的语音重组的梯度转变。

{"title":"Gestural restructuring beyond coarticulation in Korean /w/-vowel sequences: Evidence from phonetic, dialectal, and gender variation","authors":"Dae-yong Lee , Sahyang Kim , Taehong Cho","doi":"10.1016/j.wocn.2025.101456","DOIUrl":"10.1016/j.wocn.2025.101456","url":null,"abstract":"<div><div>This study examines the articulatory patterns of Korean /w/-vowel sequences by comparing tongue dorsum movement trajectories with those of corresponding plain vowels, using Electromagnetic Articulography data from 48 speakers of Seoul and North Gyeongsang dialects. The central question is whether these sequences reflect mere coarticulation or exhibit signs of gestural restructuring in the nucleus vowel. Results reveal gradient restructuring shaped by vowel constriction degree, dialect, and gender. High vowels (/wi/-/i/) show minimal divergence, mid vowels (/we/-/e/, /wɛ/-/ɛ/) moderate divergence, and low back vowels (/wa/-/a/, /wʌ/-/ʌ/) the greatest divergence—especially in dialect- and gender-specific ways. Further analysis of the /e/-/ɛ/ merger and the recent /ʌ/-/ɨ/ split in North Gyeongsang sheds light on how vowel distinctions interact with /w/. The /we/-/wɛ/ pair shows a stronger merger than /e/-/ɛ/, supporting the view that /w/ triggers gestural restructuring of the nucleus vowel and thus plays an active role in reshaping merger trajectories. This effect is further illustrated by the /wa/-/wʌ/ and /a/-/ʌ/ contrasts, with a stronger merger in the /w/-initial context—an effect notably led by male speakers. Interestingly, North Gyeongsang males preserve the /a/-/ʌ/ contrast more robustly than the /wa/-/wʌ/ contrast, possibly due to hyperarticulation of a phonetically redefined /ʌ/ resulting from the recent /ʌ/-/ɨ/ split. These findings are interpreted within a dynamical framework of gestural blending strength (GBS), which varies by vowel constriction and coarticulatory resistance but remains stable for /w/. Overall, the results suggest that what may have begun as low-level coarticulation has evolved into systematic gestural restructuring—a gradient shift toward phonological reorganization shaped by phonetic context, sound change, and sociophonetic variation.</div></div>","PeriodicalId":51397,"journal":{"name":"Journal of Phonetics","volume":"113 ","pages":"Article 101456"},"PeriodicalIF":2.4,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145424456","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"文学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Transitory and sustained Cf0 effects: Evidence from Swiss German 短暂和持续的Cf0效应：来自瑞士德语的证据

IF 2.4 1区文学 0 LANGUAGE & LINGUISTICS

Journal of Phonetics

Pub Date : 2025-10-03 DOI: 10.1016/j.wocn.2025.101453

Franka Zebe-Sheng , Camille Watter , Stephan Schmid , D. Robert Ladd

It is generally agreed that f0 following phonologically voiceless plosives is higher than after voiced plosives. Such consonant f0 (Cf0) effects have been reported in many languages. However, the phonetic basis of the ‘voiceless’ – ‘voiced’ distinction may differ between languages; for example, in English the distinction involves long-lag VOT in ‘voiceless’ plosives and short-lag VOT or prevoicing in ‘voiced’ plosives, while in Dutch the ‘voiceless’ plosives have short-lag VOT and the ‘voiced’ plosives are generally prevoiced. This study focuses on Swiss German, where neither long-lag VOT nor voicing is present: the primary difference between lenis (‘voiced’) and fortis (‘voiceless’) plosives lies in closure duration. Replicating Ladd and Schmid [Journal of Phonetics (2018), 71, 229–248], we show that both lenis and fortis plosives exhibit higher initial f0 followed by a brief fall, typical of ‘voiceless’ plosives in many languages. Using newer statistical methods (Generalised Additive Mixed Models), we also show that, during the latter part of the vowel beyond the initial f0 drop, overall f0 level is significantly higher after ‘fortis’ than after ‘lenis’ plosives. This suggests that two distinct but interacting Cf0 effects are involved. We discuss the relevance of this finding for future experimental work on Cf0.

人们普遍认为，在音系上，后面的不发音元音比后面的浊音元音要高。这种辅音f0 （Cf0）效应在许多语言中都有报道。然而，“不发声”和“发声”区分的语音基础可能因语言而异；例如，在英语中，这种区别涉及到“不发声”爆破语中的长滞后元音和“发声”爆破语中的短滞后元音或前置元音，而在荷兰语中，“不发声”爆破语有短滞后元音，而“发声”爆破语通常是前置元音。这项研究的重点是瑞士德语，那里既没有长滞后的VOT，也没有发声：lenis（“发声”）和fortis（“不发声”）爆破语的主要区别在于关闭持续时间。复制Ladd和Schmid [Journal of Phonetics(2018), 71, 229-248]，我们发现lenis和fortis爆破音都表现出更高的初始f0，然后是短暂的下降，这是许多语言中典型的“无音”爆破音。使用较新的统计方法（广义加性混合模型），我们还表明，在元音的后半部分，在初始的f0下降之后，“fortis”之后的整体f0水平明显高于“lenis”之后的爆破。这表明涉及两种不同但相互作用的Cf0效应。我们讨论了这一发现对未来Cf0实验工作的相关性。

{"title":"Transitory and sustained Cf0 effects: Evidence from Swiss German","authors":"Franka Zebe-Sheng , Camille Watter , Stephan Schmid , D. Robert Ladd","doi":"10.1016/j.wocn.2025.101453","DOIUrl":"10.1016/j.wocn.2025.101453","url":null,"abstract":"<div><div>It is generally agreed that f0 following phonologically voiceless plosives is higher than after voiced plosives. Such consonant f0 (Cf0) effects have been reported in many languages. However, the phonetic basis of the ‘voiceless’ – ‘voiced’ distinction may differ between languages; for example, in English the distinction involves long-lag VOT in ‘voiceless’ plosives and short-lag VOT or prevoicing in ‘voiced’ plosives, while in Dutch the ‘voiceless’ plosives have short-lag VOT and the ‘voiced’ plosives are generally prevoiced. This study focuses on Swiss German, where neither long-lag VOT nor voicing is present: the primary difference between lenis (‘voiced’) and fortis (‘voiceless’) plosives lies in closure duration. Replicating Ladd and Schmid [Journal of Phonetics (2018), 71, 229–248], we show that both lenis and fortis plosives exhibit higher initial f0 followed by a brief fall, typical of ‘voiceless’ plosives in many languages. Using newer statistical methods (Generalised Additive Mixed Models), we also show that, during the latter part of the vowel beyond the initial f0 drop, overall f0 level is significantly higher after ‘fortis’ than after ‘lenis’ plosives. This suggests that two distinct but interacting Cf0 effects are involved. We discuss the relevance of this finding for future experimental work on Cf0.</div></div>","PeriodicalId":51397,"journal":{"name":"Journal of Phonetics","volume":"113 ","pages":"Article 101453"},"PeriodicalIF":2.4,"publicationDate":"2025-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145221014","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"文学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Effects of minimal pair competitors on voice onset time and pitch accent production in South Swedish 最小对竞争者对瑞典南部语音开始时间和音高重音产生的影响

IF 2.4 1区文学 0 LANGUAGE & LINGUISTICS

Journal of Phonetics

Pub Date : 2025-09-05 DOI: 10.1016/j.wocn.2025.101445

Benjamin M. Kramer, Jason A. Shaw

Previous findings suggest that words in minimal pairs are hyperarticulated along the phonetic dimension that distinguishes them. We investigated the effects of minimal pair presence on the production of the pitch accent contrast and the stop voicing contrast in South Swedish; while contrastive hyperarticulation along these dimensions has been observed in other languages, these contrasts in South Swedish have a particularly low functional load and a particularly high category distance, respectively. Results from an experimental word naming task indicate that minimal pair competition does not significantly affect voice onset time in South Swedish. For the pitch accent contrast, minimal pair competition is significantly correlated with converged rather than diverged accent contours. These findings are consistent with activation dynamics of phonetic planning that are sensitive to language-specific characteristics of a contrast, such as category distance and functional load.

先前的研究结果表明，最小成对的单词在区分它们的语音维度上是高度发音的。我们研究了最小对存在对南瑞典语的音高重音对比和顿音对比产生的影响；虽然在其他语言中也观察到沿着这些维度的对比高发音，但瑞典南部的这些对比分别具有特别低的功能负荷和特别高的类别距离。实验结果表明，最小配对竞争对南瑞典语语音启动时间没有显著影响。对于音调重音对比，最小对竞争与收敛重音轮廓显著相关，而不是发散重音轮廓。这些发现与语音规划的激活动力学是一致的，语音规划对对比的特定语言特征很敏感，如类别距离和功能负荷。

引用次数: 0

The effect of rhythm on inter-gestural coupling of onset and vowel gestures and predictive timing in stuttering 节奏对口吃起音和元音手势的手势间耦合及预测时间的影响

IF 1.9 1区文学 0 LANGUAGE & LINGUISTICS

Journal of Phonetics

Pub Date : 2025-07-21 DOI: 10.1016/j.wocn.2025.101432

Mona Franke , Simone Falk , Nicole Benker , Phil Hoole

In this study we investigate articulatory timing in fluent speech production in persons who stutter (PWS) and persons who do not stutter (PWNS) by focusing on consonant–vowel (CV)-timing, which refers to the coupling of onset consonant and vowel gestures, as well as on predictive timing, which describes the synchronization of the speech onset to a rhythmic event. These two timing mechanisms are particularly interesting to investigate in relation to stuttering, given that CV-timing is especially challenging for PWS and that they exhibit differences in predictive timing related to speech-motor and manual-motor tasks, suggesting that disturbances in inter-gestural coordination and auditory-motor integration may contribute to stuttering. To shed further light on this, we examine CV-timing and predictive timing under different rhythmic conditions.

Twenty German-speaking adults (10 PWS and 10 PWNS) were recorded using electromagnetic articulography (EMA). Participants produced target words that started with a bilabial onset, followed by a vowel (/a/, /o/, or /u/) and were embedded in a carrier phrase in four different conditions: Unpaced (speaking), Tapping (speaking while concurrently tapping), Metronome (synchronizing speech to a metronome), and Metronome+Tapping (speaking to a metronome while concurrently tapping).

We found evidence for both CV-timing and predictive timing differences between PWS and PWNS. Our results suggest that in general, PWS time CV gestures closer together. However, CV-timing differences were linked to condition in an unexpected way. As to predictive timing, PWS initiated their speech later to a metronome beat than PWNS but they did not differ when timing speech to their own finger tapping, indicating that motor-pacing may stabilize the speech motor system of PWS. In the Metronome+Tapping condition, the groups appeared to rely on different rhythmic cues. While PWNS timed their speech more towards the metronome beat, PWS synchronized their speech onset closer to the finger tap. We discuss that this difference could result from differences in CV-timing. Furthermore, the potential for future research on the interplay of non-verbal and verbal motor systems and the possible benefit for the stuttering population is discussed.

在本研究中，我们研究了口吃者（PWS）和非口吃者（PWNS）在流利言语产生中的发音时间，重点关注辅音-元音（CV）时间，这是指辅音和元音的开始手势的耦合，以及预测时间，这描述了语音开始与节奏事件的同步。这两种计时机制对口吃的研究尤其有趣，因为对PWS来说，cv计时尤其具有挑战性，而且他们在言语-运动和手动-运动任务相关的预测计时方面表现出差异，这表明手势间协调和听觉-运动整合的障碍可能导致口吃。为了进一步阐明这一点，我们研究了不同节奏条件下的cv计时和预测计时。使用电磁关节造影（EMA）对20名讲德语的成年人（10名PWS和10名PWNS）进行了记录。参与者提出的目标词以双音节开头，后面跟着一个元音（/a/， /o/或/u/），并在四种不同的条件下嵌入载体短语：无节奏（说话），敲击（说话时同时敲击），节拍器（将语音与节拍器同步），节拍器+敲击（同时敲击节拍器说话）。我们发现了PWS和PWNS之间cv时间和预测时间差异的证据。我们的研究结果表明，在一般情况下，PWS时间CV手势更接近。然而，cv时间的差异以一种意想不到的方式与病情联系在一起。在预测计时方面，PWS比PWNS晚于节拍器节拍启动言语，但在将言语计时为自己的手指轻敲时，两者没有差异，说明运动起搏可能稳定了PWS的言语运动系统。在节拍器+敲击条件下，各组似乎依赖于不同的节奏线索。PWNS的说话时间更接近节拍器的节拍，而PWS的说话时间更接近手指的敲击。我们讨论了这种差异可能是由于cv时间的差异造成的。此外，本文还讨论了未来研究非语言和语言运动系统相互作用的潜力以及对口吃人群可能带来的好处。

{"title":"The effect of rhythm on inter-gestural coupling of onset and vowel gestures and predictive timing in stuttering","authors":"Mona Franke , Simone Falk , Nicole Benker , Phil Hoole","doi":"10.1016/j.wocn.2025.101432","DOIUrl":"10.1016/j.wocn.2025.101432","url":null,"abstract":"<div><div>In this study we investigate articulatory timing in fluent speech production in persons who stutter (PWS) and persons who do not stutter (PWNS) by focusing on consonant–vowel (CV)-timing, which refers to the coupling of onset consonant and vowel gestures, as well as on predictive timing, which describes the synchronization of the speech onset to a rhythmic event. These two timing mechanisms are particularly interesting to investigate in relation to stuttering, given that CV-timing is especially challenging for PWS and that they exhibit differences in predictive timing related to speech-motor and manual-motor tasks, suggesting that disturbances in inter-gestural coordination and auditory-motor integration may contribute to stuttering. To shed further light on this, we examine CV-timing and predictive timing under different rhythmic conditions.</div><div>Twenty German-speaking adults (10 PWS and 10 PWNS) were recorded using electromagnetic articulography (EMA). Participants produced target words that started with a bilabial onset, followed by a vowel (/a/, /o/, or /u/) and were embedded in a carrier phrase in four different conditions: Unpaced (speaking), Tapping (speaking while concurrently tapping), Metronome (synchronizing speech to a metronome), and Metronome+Tapping (speaking to a metronome while concurrently tapping).</div><div>We found evidence for both CV-timing and predictive timing differences between PWS and PWNS. Our results suggest that in general, PWS time CV gestures closer together. However, CV-timing differences were linked to condition in an unexpected way. As to predictive timing, PWS initiated their speech later to a metronome beat than PWNS but they did not differ when timing speech to their own finger tapping, indicating that motor-pacing may stabilize the speech motor system of PWS. In the Metronome+Tapping condition, the groups appeared to rely on different rhythmic cues. While PWNS timed their speech more towards the metronome beat, PWS synchronized their speech onset closer to the finger tap. We discuss that this difference could result from differences in CV-timing. Furthermore, the potential for future research on the interplay of non-verbal and verbal motor systems and the possible benefit for the stuttering population is discussed.</div></div>","PeriodicalId":51397,"journal":{"name":"Journal of Phonetics","volume":"112 ","pages":"Article 101432"},"PeriodicalIF":1.9,"publicationDate":"2025-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144679165","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"文学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Phonetic information in the vowel spectrum: the meaning of mel-Frequency Cepstral Coefficients 元音谱中的语音信息：mel-Frequency倒谱系数的意义

IF 1.9 1区文学 0 LANGUAGE & LINGUISTICS

Journal of Phonetics

Pub Date : 2025-07-17 DOI: 10.1016/j.wocn.2025.101434

Khalil Iskarous , Alessandro Vietti

There is still disagreement in the acoustic phonetics literature on how phonetic information is encoded in the vowel acoustic spectrum. The “formant hypothesis” holds that formant frequency locations are the primary encoding of phonetic information. But perceptual experiments have shown that listeners can identify vowels, to a certain extent, even when formant peaks are suppressed. This has given rise to the “whole-spectrum” hypothesis, which describes each vowel segment in terms of a high-dimensional description of its entire spectrum. While the “whole-spectrum” hypothesis better predicts suppressed-formant vowel perception, one advantage of the “formant hypothesis” is that it parameterizes a vowel inventory of a language in terms of featural classes indexed by a few values of formant frequencies. These frequency scales serve to describe a language’s phonological organization and sound change. In this paper, we show that the mel-frequency Cepstral Coefficients (MFCCs), whole-spectrum parameterizations that have been used in speech technology from the 1970’s till today, also have a phonetic interpretation leading to the same featural classes as traditional description. This is despite the fact that for many decades they have been thought to not be interpretable. Our arguments are based on analyses of all vowel data from the TIMIT database, with large amounts of speaker, context, prosodic, and dialectal variability, using information theory, effect-size statistics, and Fourier theory. Our goal is to show that MFCCs can be useful for further developments in the field of acoustic phonetics, because while they extract phonetically-distinctive information from the entire spectrum, they can also further understanding of the linguistic structure of vowel spaces.

语音信息如何在元音声谱中编码，在声学语音学文献中仍存在分歧。“共振峰假说”认为共振峰频率位置是语音信息的主要编码。但感知实验表明，即使在形成峰被抑制的情况下，听众也能在一定程度上识别元音。这就产生了“全谱”假说，它用整个谱的高维描述来描述每个元音片段。虽然“全谱”假说能更好地预测被抑制的形成峰元音感知，但“形成峰假说”的一个优点是，它以几个形成峰频率值为索引的特征类别来参数化语言的元音清单。这些频率尺度用来描述一种语言的语音组织和声音变化。在本文中，我们展示了mel-frequency倒谱系数（MFCCs），即从20世纪70年代至今一直用于语音技术的全频谱参数化，也具有语音解释，导致与传统描述相同的特征类别。尽管几十年来它们一直被认为是不可解释的。我们的论点是基于对TIMIT数据库中所有元音数据的分析，使用信息论、效应大小统计和傅立叶理论，分析了大量的说话人、上下文、韵律和方言差异。我们的目标是证明mfcc在声学语音学领域的进一步发展是有用的，因为当它们从整个频谱中提取语音特征信息时，它们也可以进一步理解元音空间的语言结构。

{"title":"Phonetic information in the vowel spectrum: the meaning of mel-Frequency Cepstral Coefficients","authors":"Khalil Iskarous , Alessandro Vietti","doi":"10.1016/j.wocn.2025.101434","DOIUrl":"10.1016/j.wocn.2025.101434","url":null,"abstract":"<div><div>There is still disagreement in the acoustic phonetics literature on how phonetic information is encoded in the vowel acoustic spectrum. The “formant hypothesis” holds that formant frequency locations are the primary encoding of phonetic information. But perceptual experiments have shown that listeners can identify vowels, to a certain extent, even when formant peaks are suppressed. This has given rise to the “whole-spectrum” hypothesis, which describes each vowel segment in terms of a high-dimensional description of its entire spectrum. While the “whole-spectrum” hypothesis better predicts suppressed-formant vowel perception, one advantage of the “formant hypothesis” is that it parameterizes a vowel inventory of a language in terms of featural classes indexed by a few values of formant frequencies. These frequency scales serve to describe a language’s phonological organization and sound change. In this paper, we show that the mel-frequency Cepstral Coefficients (MFCCs), whole-spectrum parameterizations that have been used in speech technology from the 1970’s till today, also have a phonetic interpretation leading to the same featural classes as traditional description. This is despite the fact that for many decades they have been thought to not be interpretable. Our arguments are based on analyses of all vowel data from the TIMIT database, with large amounts of speaker, context, prosodic, and dialectal variability, using information theory, effect-size statistics, and Fourier theory. Our goal is to show that MFCCs can be useful for further developments in the field of acoustic phonetics, because while they extract phonetically-distinctive information from the entire spectrum, they can also further understanding of the linguistic structure of vowel spaces.</div></div>","PeriodicalId":51397,"journal":{"name":"Journal of Phonetics","volume":"112 ","pages":"Article 101434"},"PeriodicalIF":1.9,"publicationDate":"2025-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144655552","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"文学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Effects of sentential context on nonnative recognition of reduced speech: Does meaning explain it all? 句子语境对非母语语音简化识别的影响：意义能解释一切吗？

IF 1.9 1区文学 0 LANGUAGE & LINGUISTICS

Journal of Phonetics

Pub Date : 2025-07-15 DOI: 10.1016/j.wocn.2025.101433

Bihua Chen , Isabelle Darcy

In casual speech, reduction of segments or even syllables is common. Native (L1) listeners recover these reduced forms by recruiting not only semantic and syntactic but also fine-grained acoustic cues in the surrounding utterance. Whether second-language (L2) listeners exploit the same constellation of cues is still poorly understood. We therefore compared 21 L1 English listeners and 21 Mandarin learners of English as they identified reduced targets (e.g., /tuɪnə/ ‘too into’) presented in one of three contexts: Isolation, Textual (orthographic context only), and Auditory (orthography plus the original carrier sentence). Accuracy patterns revealed a graded facilitation hierarchy. For L2 listeners, semantic-syntactic information alone (Textual) boosted recognition relative to Isolation, and adding acoustic context produced a further significant gain. Nevertheless, both effects were smaller for L2 than for L1 listeners, indicating less effective contextual integration in the L2 processing mechanism. The findings refine accounts of reduced-speech perception by showing that L2 listeners can harness acoustic context, but less efficiently than L1 listeners.

在随意的讲话中，省略音段甚至音节是很常见的。母语（L1）听者不仅通过收集周围话语中的语义和句法线索，而且还通过收集细粒度的声音线索来恢复这些简化形式。第二语言（L2）的听者是否也利用了同样的线索群，我们仍然知之甚少。因此，我们比较了21名母语英语听众和21名普通话英语学习者，因为他们识别了在三种语境中出现的减少目标（例如，/tu / n / / ' too into '）：孤立、文本（仅限正字法语境）和听觉（正字法加上原始载体句）。准确度模式显示了一个分级的促进层次。对于二语听者来说，单独的语义句法信息（文本信息）相对于孤立信息提高了识别能力，而添加声学上下文则产生了进一步的显著增益。然而，这两种影响对于第二语言的听者来说都比母语听者要小，这表明第二语言加工机制中的语境整合效果较差。研究结果表明，第二语言的听者可以利用声音环境，但效率不如第一语言的听者，从而完善了对言语感知减少的解释。

{"title":"Effects of sentential context on nonnative recognition of reduced speech: Does meaning explain it all?","authors":"Bihua Chen , Isabelle Darcy","doi":"10.1016/j.wocn.2025.101433","DOIUrl":"10.1016/j.wocn.2025.101433","url":null,"abstract":"<div><div>In casual speech, reduction of segments or even syllables is common. Native (L1) listeners recover these reduced forms by recruiting not only semantic and syntactic but also fine-grained acoustic cues in the surrounding utterance. Whether second-language (L2) listeners exploit the same constellation of cues is still poorly understood. We therefore compared 21 L1 English listeners and 21 Mandarin learners of English as they identified reduced targets (e.g., /tuɪnə/ ‘too into’) presented in one of three contexts: Isolation, Textual (orthographic context only), and Auditory (orthography plus the original carrier sentence). Accuracy patterns revealed a graded facilitation hierarchy. For L2 listeners, semantic-syntactic information alone (Textual) boosted recognition relative to Isolation, and adding acoustic context produced a further significant gain. Nevertheless, both effects were smaller for L2 than for L1 listeners, indicating less effective contextual integration in the L2 processing mechanism. The findings refine accounts of reduced-speech perception by showing that L2 listeners can harness acoustic context, but less efficiently than L1 listeners.</div></div>","PeriodicalId":51397,"journal":{"name":"Journal of Phonetics","volume":"112 ","pages":"Article 101433"},"PeriodicalIF":1.9,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144632575","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"文学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Cascading activation in spoken word production drives incomplete neutralization: An internet-based study of Mandarin 3rd tone sandhi 口语单词产生中的级联激活驱动不完全中和：基于网络的普通话三变调研究

IF 1.9 1区文学 0 LANGUAGE & LINGUISTICS

Journal of Phonetics

Pub Date : 2025-07-12 DOI: 10.1016/j.wocn.2025.101428

Yuyu Zeng , Chang Wang , Jie Zhang

Incomplete neutralization occurs when two underlying contrastive sounds are phonologically neutralized but remain phonetically distinct (e.g., “latter” and “ladder” become homophonous when the intervocalic stops are flapped in American English). Its proper understanding is foundational to phonology and speech production. Using the incomplete neutralization of the Mandarin 3rd tone sandhi as a test case (T3 + T3 → T2 + T3), we confirmed the presence of this incomplete neutralization with generalized additive modeling (GAMM) and growth curve analysis (GCA). Crucially, we found that the two tones (T2 and T3) became more neutralized when speakers were additionally required to perform a concurrent verbal working memory task while speaking; similar patterns were found when pseudowords were tested, although the overall effects were weaker. Since the concurrent verbal working memory task is expected to add processing load and decrease cascading activation in the spoken word production process, our results suggest that cascading activation, which permits upstream distinctions to surface in downstream acoustics, drives incomplete neutralization. Our study shows how embracing cascading activation can inform the long-standing debate between discrete vs. exemplar representations/operations surrounding incomplete neutralization. How cascading activation is compatible with the core assumptions of generative phonology is also discussed.

不完全中和发生在两个潜在的对比音在语音上中和，但在语音上仍然不同的情况下（例如，在美式英语中，“later”和“ladder”在中间停顿时成为同音音）。正确理解它是音韵学和语音生成的基础。以普通话三调连调的不完全中和为例（T3 + T3→T2 + T3），利用广义加性建模（GAMM）和生长曲线分析（GCA）证实了这种不完全中和的存在。至关重要的是，我们发现当说话者在说话时被要求同时执行言语工作记忆任务时，这两个音调（T2和T3）变得更加中立；在测试假词时也发现了类似的模式，尽管总体效果较弱。由于并发言语工作记忆任务预计会增加加工负荷并减少口语单词产生过程中的级联激活，因此我们的研究结果表明，级联激活允许上游声学差异在下游声学中浮出水面，从而驱动不完全中和。我们的研究表明，拥抱级联激活可以为围绕不完全中和的离散与范例表示/操作之间的长期争论提供信息。级联激活如何与生成音韵学的核心假设兼容也进行了讨论。

{"title":"Cascading activation in spoken word production drives incomplete neutralization: An internet-based study of Mandarin 3rd tone sandhi","authors":"Yuyu Zeng , Chang Wang , Jie Zhang","doi":"10.1016/j.wocn.2025.101428","DOIUrl":"10.1016/j.wocn.2025.101428","url":null,"abstract":"<div><div>Incomplete neutralization occurs when two underlying contrastive sounds are phonologically neutralized but remain phonetically distinct (e.g., “latter” and “ladder” become homophonous when the intervocalic stops are flapped in American English). Its proper understanding is foundational to phonology and speech production. Using the incomplete neutralization of the Mandarin 3rd tone sandhi as a test case (T3 + T3 → T2 + T3), we confirmed the presence of this incomplete neutralization with generalized additive modeling (GAMM) and growth curve analysis (GCA). Crucially, we found that the two tones (T2 and T3) became more neutralized when speakers were additionally required to perform a concurrent verbal working memory task while speaking; similar patterns were found when pseudowords were tested, although the overall effects were weaker. Since the concurrent verbal working memory task is expected to add processing load and decrease cascading activation in the spoken word production process, our results suggest that cascading activation, which permits upstream distinctions to surface in downstream acoustics, drives incomplete neutralization. Our study shows how embracing cascading activation can inform the long-standing debate between discrete vs. exemplar representations/operations surrounding incomplete neutralization. How cascading activation is compatible with the core assumptions of generative phonology is also discussed.</div></div>","PeriodicalId":51397,"journal":{"name":"Journal of Phonetics","volume":"112 ","pages":"Article 101428"},"PeriodicalIF":1.9,"publicationDate":"2025-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144605212","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"文学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Perceived cross-linguistic similarity of retroflexes in trilingual, bilingual and native listener groups 三语、双语和母语听者群体中反旋音的感知跨语言相似性

IF 1.9 1区文学 0 LANGUAGE & LINGUISTICS

Journal of Phonetics

Pub Date : 2025-07-11 DOI: 10.1016/j.wocn.2025.101429

Anna Balas , Krzysztof Hwaszcz , Kamil Kaźmierski , Magdalena Wrembel

This paper investigates the perceived cross-linguistic similarity of retroflexes from a broad multilingual perspective by employing trilingual and bilingual learners and native users as three distinct listener groups. Previous research has demonstrated that L2 learners rely on their L1 in non-native speech perception. However, no study has examined how L3 learners perceive differences between retroflex sounds in their L1, L2, and L3. In a series of three parallel studies, we examined cross-linguistic similarity of Norwegian retroflexes and similar retroflex and non-retroflex sounds by trilingual (L1 Polish, L2 English and L3 Norwegian), bilingual (L1 Polish, L2 English) and Norwegian control (L1 Norwegian, L2 English) listeners. The listeners assessed similarity between the Norwegian and Polish or English sounds primarily based on the place and manner of articulation rather than retroflexion. The results, where condition specifies the presence or absence of agreement in terms of retroflexion and place/manner of articulation, demonstrated that all the two-way interactions: condition:language, condition:group, language:group and the three-way interaction were significant. The study revealed that experience with a given language did not influence similarity ratings in a wholesale manner but rather in a precise manner related to the presence or absence of retroflexion. The results also showed that the perceived cross-linguistic similarity by multilinguals was gradient in nature. The study provides new insights into research on the perception of retroflexes and multilingual perception by participants differing in the amount of experience with the languages of the stimuli: from L1 controls through L2 and L3 learners to naïve listeners.

本文将三语和双语学习者和母语使用者作为三个不同的听者群体，从广泛的多语视角研究了反旋的跨语言相似性感知。先前的研究表明，第二语言学习者在非母语语言感知方面依赖于他们的母语。然而，没有研究考察过三语学习者如何感知L1、L2和L3中反音的差异。在一系列的三个平行研究中，我们对三语（第一语言波兰语、第二语言英语和第三语言挪威语）、双语（第一语言波兰语、第二语言英语）和挪威对照（第一语言挪威语、第二语言英语）的听者进行了跨语言相似性的研究。听众评估挪威语和波兰语或英语发音的相似度主要基于发音的位置和方式，而不是反音。结果表明，条件指定了在反折和发音位置/方式方面是否存在一致，所有的双向互动：条件：语言，条件：组，语言：组和三方互动都是显著的。研究表明，使用特定语言的经历并没有全面地影响相似度评分，而是以一种与有无反义相关的精确方式影响相似度评分。结果还表明，多语者感知的跨语言相似性具有梯度特征。该研究为不同刺激物语言经验的参与者对反射和多语言感知的研究提供了新的见解：从L1控制者到L2和L3学习者再到naïve听者。

{"title":"Perceived cross-linguistic similarity of retroflexes in trilingual, bilingual and native listener groups","authors":"Anna Balas , Krzysztof Hwaszcz , Kamil Kaźmierski , Magdalena Wrembel","doi":"10.1016/j.wocn.2025.101429","DOIUrl":"10.1016/j.wocn.2025.101429","url":null,"abstract":"<div><div>This paper investigates the perceived cross-linguistic similarity of retroflexes from a broad multilingual perspective by employing trilingual and bilingual learners and native users as three distinct listener groups. Previous research has demonstrated that L2 learners rely on their L1 in non-native speech perception. However, no study has examined how L3 learners perceive differences between retroflex sounds in their L1, L2, and L3. In a series of three parallel studies, we examined cross-linguistic similarity of Norwegian retroflexes and similar retroflex and non-retroflex sounds by trilingual (L1 Polish, L2 English and L3 Norwegian), bilingual (L1 Polish, L2 English) and Norwegian control (L1 Norwegian, L2 English) listeners. The listeners assessed similarity between the Norwegian and Polish or English sounds primarily based on the place and manner of articulation rather than retroflexion. The results, where condition specifies the presence or absence of agreement in terms of retroflexion and place/manner of articulation, demonstrated that all the two-way interactions: condition:language, condition:group, language:group and the three-way interaction were significant. The study revealed that experience with a given language did not influence similarity ratings in a wholesale manner but rather in a precise manner related to the presence or absence of retroflexion. The results also showed that the perceived cross-linguistic similarity by multilinguals was gradient in nature. The study provides new insights into research on the perception of retroflexes and multilingual perception by participants differing in the amount of experience with the languages of the stimuli: from L1 controls through L2 and L3 learners to naïve listeners.</div></div>","PeriodicalId":51397,"journal":{"name":"Journal of Phonetics","volume":"112 ","pages":"Article 101429"},"PeriodicalIF":1.9,"publicationDate":"2025-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144605211","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"文学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Velar palatalization, phonologization, and sound change – A comparative acoustic study of /k/-fronting in Majorcan Catalan 马略坎加泰罗尼亚语/k/前音的比较声学研究

IF 1.9 1区文学 0 LANGUAGE & LINGUISTICS

Journal of Phonetics

Pub Date : 2025-07-10 DOI: 10.1016/j.wocn.2025.101430

Miquel Simonet , Marta Ramírez Martínez , Francesc Torres-Tamarit

This study explores the acoustics of velar palatalization in two subvarieties of Majorcan Catalan, Manacor (palatalizing) and Artà (nonpalatalizing). Three production studies are reported: i) a study of /k/-fronting in the context of front, central, and back vowels; ii) a study of /a/-fronting in the context of /k/ and /p/; and iii) a study of /k/-fronting in various vowel contexts in the participants’ L2, Spanish. First, while we captured /k/-fronting in the progression /o/ > /a ə/ > /i/ in both subvarieties, effect sizes were much larger in Manacor than in Artà. There were no group differences in the acoustics of /k/ in the context of the back vowel, but there were large differences in the other vowel contexts, particularly before the central vowels. We postulate that, whereas the degree of palatalization found in Artà may result from universal coarticulatory principles, palatalization in Manacor results from speaker-controlled phonetic behavior: enhanced coarticulation. Second, we found that in Manacor (but not Artà) /a/ was more fronted when it followed /k/ that when it followed /p/. We suggest that the /a/-fronting pattern found in Manacor results from the influence of its velar-palatalization process and not vice versa. Finally, we found that the enhanced velar-palatalization process in the Manacor sample was not transferred to their L2. We discuss the implications of our conclusion for our understanding of the diachrony of velar palatalization in Romance.

本研究探讨了马略坎加泰罗尼亚语的两个亚种，马纳科尔（舌化）和阿尔托孔（非舌化）的舌化声学。报告了三个生产研究：i)在前、中、后元音语境中/k/-fronting的研究；Ii)在/k/和/p/上下文中对/a/-前置的研究；iii)研究参与者在西班牙语第二语言中不同元音语境中/k/的前置。首先，当我们捕捉到/k/-在/o/ >；/a æ / >；/i/在这两个亚品种中，马纳科尔的效应量远大于阿尔托姆。在后面元音/k/的发音上下文中，没有组间差异，但在其他元音上下文中，特别是在中心元音之前，有很大的差异。我们假设，尽管在art中发现的舌化程度可能是由普遍的协同发音原则造成的，但在Manacor中发现的舌化是由说话人控制的语音行为造成的：增强的协同发音。其次，我们发现在Manacor（而不是art）中，/a/跟在/k/后面比跟在/p/后面更靠前。我们认为在Manacor发现的/a/-前缘模式是由于其腭腭化过程的影响，而不是相反。最后，我们发现Manacor样品中增强的腭腭化过程并没有转移到L2。我们讨论了我们的结论对我们理解《浪漫》中腭腭化的历时性的意义。

{"title":"Velar palatalization, phonologization, and sound change – A comparative acoustic study of /k/-fronting in Majorcan Catalan","authors":"Miquel Simonet , Marta Ramírez Martínez , Francesc Torres-Tamarit","doi":"10.1016/j.wocn.2025.101430","DOIUrl":"10.1016/j.wocn.2025.101430","url":null,"abstract":"<div><div>This study explores the acoustics of velar palatalization in two subvarieties of Majorcan Catalan, Manacor (palatalizing) and Artà (nonpalatalizing). Three production studies are reported: i) a study of /k/-fronting in the context of front, central, and back vowels; ii) a study of /a/-fronting in the context of /k/ and /p/; and iii) a study of /k/-fronting in various vowel contexts in the participants’ L2, Spanish. First, while we captured /k/-fronting in the progression /o/ > /a ə/ > /i/ in both subvarieties, effect sizes were much larger in Manacor than in Artà. There were no group differences in the acoustics of /k/ in the context of the back vowel, but there were large differences in the other vowel contexts, particularly before the central vowels. We postulate that, whereas the degree of palatalization found in Artà may result from universal coarticulatory principles, palatalization in Manacor results from speaker-controlled phonetic behavior: enhanced coarticulation. Second, we found that in Manacor (but not Artà) /a/ was more fronted when it followed /k/ that when it followed /p/. We suggest that the /a/-fronting pattern found in Manacor results from the influence of its velar-palatalization process and not <em>vice versa</em>. Finally, we found that the enhanced velar-palatalization process in the Manacor sample was not transferred to their L2. We discuss the implications of our conclusion for our understanding of the diachrony of velar palatalization in Romance.</div></div>","PeriodicalId":51397,"journal":{"name":"Journal of Phonetics","volume":"112 ","pages":"Article 101430"},"PeriodicalIF":1.9,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144596707","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"文学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0