Journal of Phonetics最新文献

A common goal of kinematic studies on disordered speech is the identification of speech motor impairments that negatively impact speech function. Although it is well-known that the kinematic contours of speakers with speech disorders often deviate considerably from those of neurotypical speakers, systematic quantitative assessments of these impairment-related movement disturbances remain challenging. Kinematic measurement approaches are commonly grounded in models and theories that have emerged exclusively from observations of healthy speakers. However, often these models cannot accommodate the deviant articulatory behaviors of speakers with speech motor impairment. In the present paper, we address this problem. By considering noise as a factor in Articulatory Phonology/Task Dynamics (AP/TD), we can account for articulatory behaviors that are known to occur in healthy speakers (e.g., during slow speech) as well as in speakers with motor speech impairments. In a proof of concept, we descriptively compare modeled articulatory behaviors that include noise at various levels with empirical data. We view such an extension of the AP/TD as a first step towards a more comprehensive speech production model that can serve as a theoretical framework to study the speech production mechanism in healthy speakers and speakers with motor speech impairments.

Creaky voice is a non-modal voice quality generally described as sounding pulse-like and low in pitch. While empirical studies have produced mixed results when it comes to creak prevalence by speaker gender, creaky voice is stereotypically associated with women’s speech. Past research has investigated whether listeners are facilitated in their identification of creaky voice through the degree of pitch differences between modal and creaky voice or by social biases associating creak with women’s speech. Results, however, have been relatively inconclusive. The present study addresses this question through a perception experiment, using highly controlled stimuli. 258 listeners were asked to identify whether they heard creak or not when presented with stimuli manipulated for f0 and creaky voice from two-word phrases produced by a male and a female speaker of Australian English. Accuracy data and response times were analysed. Findings suggest that speakers of Australian English rely less on social biases when identifying creaky voice and instead make decisions based on pitch, modulated by their experience-based expectations about typical pitch ranges according to speaker gender. Results emphasise the importance of incorporating characteristics of the speaker into models of the perception of creaky voice, and voice quality more generally.

This study investigates cue-weighting differences in intonation perception between tonal and non-tonal languages, specifically focusing on how native Spanish listeners and Mandarin learners of Spanish identify intonation categories using changes in multiple acoustic dimensions. Employing a relatively continuous response scale, we analyzed listener performance in two perceptual tests, in which stimuli were generated by manipulating the suprasegmental cues in sentence-final positions. The results of data analyses indicate that while f0 and duration cues are significant for intonation categorization in both Spanish and Mandarin listeners, intensity appears to be a redundant cue that exerts limited effect only on native Spanish listeners. Contrary to the general claim of a tonal language benefit in pitch perception, our two language groups showed similar sensitivities to f0 linear transitions perceived as sentence intonation. Moreover, Spanish natives used higher f0 contours for question recognition compared to Mandarin learners and relied more heavily on secondary cues in their auditory judgments. The study also demonstrates that perceptual weighting varies across acoustic conditions and stress patterns, suggesting that the dynamic mapping between acoustics and intonation is shaped by language background as well as specific acoustic and word-level suprasegmental contexts.

Lombard speech, speech produced in noise, has been extensively studied in native speakers, while non-native Lombard speech research is limited. This article presents the first corpus of non-native Lombard speech, the Dutch English Lombard Native Non-Native corpus, which includes plain and Lombard read speech from native American-English, non-native English (native Dutch), and native Dutch women. The location of contrastive focus is systematically varied in the sentences. We investigated how intensity, spectral center of gravity, word duration, and VOT varies in the corpus as a function of plain versus Lombard speech and whether it is modulated by the speaker’s nativeness and of the language. We did not find differences in how the native and non-native English speakers adapted their English speech in noise, indicating that the Dutch non-native speakers produced Lombard speech similarly to the native English. The comparison of the native Dutch and non-native English sentences produced by the same participants nevertheless suggests that, for all acoustic measurements except word duration, the Dutch speakers adapted their Lombard speech differently in native Dutch than in non-native English. Combined, this would indicate that, when speaking English, Dutch speakers adapt their way of speaking in noise to the way native English speakers do.

This paper presents the first systematic acoustic analysis of a rhotic accent in present-day England. The dataset comprises spontaneous and elicited speech of 28 speakers from Blackburn in Lancashire, Northern England, where residual rhoticity remains, having never been lost in the earlier sound change which rendered most of England non-rhotic. Although sociolinguistic studies of rhoticity in England exist, we have almost no description of its phonetic properties. Moreover, most sociolinguistic studies focus on the South West of England and relatively little is known about rhoticity in the North. Our study is timely because Northern rhoticity is predicted to disappear in the next few generations, a process which is now complete in many areas of the South West. Our results demonstrate that rhoticity is still present in Blackburn, although non-prevocalic /r/ is weaker when compared to other rhotic varieties of English such as those in Scotland and North America. We find that non-prevocalic /r/ is phonetically weakening in apparent time, with the F3-F2 difference being larger for younger speakers as well as females. We present additional social and linguistic factors affecting its potential demise, and discuss how our results contribute to our understanding of historical /r/-loss in Anglo-English.

The two languages of a bilingual speaker are interconnected and mutually influence linguistic forms and structures. This study presents a case in which two languages in contact exhibit phonotactic asymmetries but converge on abstract phonological units by bilingual speakers. The specific case examined here concerns the change-in-progress of unreleased final stops among young Mandarin-Min bilingual speakers in Taiwan. Phonotactically, obstruent finals are illegal in Taiwan Mandarin, whereas Taiwanese Southern Min (TSM), a local substratum language, allows obligatorily unreleased final stops. In the discrimination of stimuli modeled after TSM, bilingual listeners were consistently outperformed by Korean listeners, a non-native reference group without restrictions against obstruent finals. A follow-up production study revealed that final stops produced by the bilingual speakers were prone to deletion accompanied by vowel lengthening, similar to a long vowel in an open syllable, as well as frequent substitution. Furthermore, strong correlations were found between bilingual speakers’ perception and production accuracy, indicating a bidirectional co-evolution between perception and production during language development. Taken together, the results suggest that a loss of unreleased final stops is underway in TSM through the structural convergence of two interacting phonological systems within bilingual individuals.

Previous research has suggested a relationship between musical abilities and the perception of speech prosody. However, effect sizes and significance differ across studies. In a meta-analysis, we assessed the overall size of this relation across 109 studies and investigated which factors moderated the effect. We found a significant, medium-sized positive correlation between musical abilities and speech prosody perception. This correlation was larger for studies on non-native compared to native prosody perception. We attribute this difference to ceiling performance in native perception, while non-native perception may be more difficult and can thus be facilitated by musical abilities. In addition, prosody perception was more strongly correlated with music perception than with music training, possibly because training metrics disregard untrained individuals with naturally strong musical abilities. Further analyses showed a stronger correlation for prosodic pitch compared to prosodic timing perception, and a stronger correlation for behavioural accuracy measures compared to reaction times. We did not find differences in effects between linguistic and emotional prosody, between L1 tone language users or non-tone language users, or between adults and children. This meta-analysis generally supports theories proposing a connection between music and speech prosody. Furthermore, this study highlights the potential importance of individuals’ musical abilities for the acquisition of second language prosody.

Phonetic imitation has been found to be mediated by phonological contrast. For features whose values vary around a phonological prototype, the imitation is distorted by the phonological category, i.e., the imitation is nonlinear. This phonological mediation effect was mostly found in segmental features such as VOT and formants. Supra-segmental features, on the contrary, are generally found to be easy to imitate, i.e., the imitation is linear. Nevertheless, whether the phonological effect exists in the imitation of supra-segmental features is not fully understood. This study, through an imitation experiment of Mandarin flat-falling tonal continua, examined whether a supra-segmental feature would be linearly imitated when it is the primary cue (F0 range) and the non-primary cue (duration) to the tonal contrast, respectively. Results showed that F0 range imitation was non-linear while duration imitation was linear. This reveals that the phonological effect is stronger in mediating imitation than would be predicted by the general hypothesis that supra-segmental features are easier to imitate.

How do listeners understand what they are hearing? Humans hearing speech perform spoken word recognition, recognizing what words they are hearing in a speech stream in order to understand the meaning. Phonetics refers to the properties of the speech at a detailed level, particularly below the level of segmental phonemic distinctions. In order to recognize spoken words, listeners have to extract information from the detailed acoustic signal in some way, but theories differ about whether listeners extract phonemes, whole words, or other units, by what mechanism, and they differ on what kinds of information are stored in the lexicon. The process of spoken word recognition can be affected by any number of situations such as the speaker or listener being a non-native of the language or dialect, being a child, having a speech/hearing disability, hearing speech in noise, the speech itself containing variability, or many other situations. Any of these situations can shed light on theoretical questions by giving a fuller picture of how listeners recognize words. This chapter examines what we have learned in these first ∼21 years of the 21st century about how phonetics interacts with spoken word recognition.

This is the first study to examine stop voicing perception in the societal (English) and heritage language (Spanish) of bilingual preschoolers. The study a) compares bilinguals’ English perception patterns to those of monolinguals; b) it examines how child-internal (age) and external variables (input quantity and input diversity) predict English and Spanish perceptual performance; and c) it compares bilinguals’ perception patterns across languages. Perception was assessed through a forced-choice minimal-pair identification task in which children heard synthesized audio stimuli that varied systematically along a /p-b/ and /t-d/ Voice Onset Time (VOT) continuum and were asked to match them with one of two pictures for each contrast. The results of Bayesian mixed-effects logistic regression analyses indicate that the bilinguals’ category boundary for English stops was impacted by their experience with Spanish, with more short-lag VOT tokens being perceived as voiceless consistent with Spanish VOT. Age solely predicted English perceptual skills, whereas input quantity was the only moderator of Spanish perceptual performance. Finally, the bilingual children showed separate stop voicing contrasts in each language, although perceptual performance was already more mature in English by preschool age. Implications for theories of bilingual speech learning and the role of sociolinguistic variables are discussed.