Phonetic Context Research Articles

Comparing non-native and native speech: Are L2 productions more variable?

Foreign-accented speech of second language learners is often difficult to understand for native listeners of that language. Part of this difficulty has been hypothesized to be caused by increased within-category variability of non-native speech. However, until recently, there have been few direct tests for this hypothesis. The realization of vowels and word-final stops in productions of native-English L1 speakers and native-Mandarin speakers of L2 English is compared. With the largest sample size to date, it is shown that at least proficient non-native speakers exhibit little or no difference in category variability compared to native speakers. This is shown while correcting for the effects of phonetic context. The same non-native speakers show substantial deviations from native speech in the central tendencies (means) of categories, as well as in the correlations among cues they produce. This relativizes a common and a priori plausible assumption that competition between first and second language representations necessarily leads to increased variability-or, equivalently, decreased precision, consistency, and stability-of non-native speech. Instead, effects of non-nativeness on category variability are category- and cue-specific.

Loose lips and tongue tips: The central role of the /r/-typical labial gesture in Anglo-English

This paper presents acoustic and articulatory data from prevocalic /r/ in the non-rhotic variety of English spoken in England, Anglo-English. Although traditional descriptions suggest that Anglo-English /r/ is produced using a tip-up tongue configuration, ultrasound data from 24 speakers show similar patterns of lingual variation to those reported in rhotic varieties, with a continuum of possible tongue shapes from bunched to retroflex. However, the number of Anglo-English speakers using exclusively tip-up variants is higher than that reported in American English across all phonetic contexts. It is generally agreed that English /r/ may be labialised, but the exact contribution of the lips has yet to be explored. Lip camera data reveal significantly more lip protrusion in bunched tongue configurations than retroflex ones. These results indicate that the differing degrees of lip protrusion may contribute to maintaining a stable acoustic output across the different tongue shapes. An articulatory-acoustic trading relation between the sublingual space and the degree of lip protrusion is proposed. Finally, we suggest that Anglo-English /r/ has a specific lip posture which differs from that of /w/. We relate the development of such a posture to Anglo-English speakers’ exposure to labiodental variants and to the pressure to maintain a perceptual contrast between /r/ and /w/.

Lowered F2 observed in uvular rhotics involves a tongue root gesture: Evidence from Upper Sorbian.

Upper Sorbian, an endangered West Slavic language spoken in Germany, is unusual among Slavic languages in having a uvular rhotic /ʀ/. This paper focuses on the gestural configuration and coarticulatory resistance of the uvular rhotic and explores the relation between the articulation and acoustics of this sound. Ultrasound tongue imaging data were collected from six native speakers of Upper Sorbian, who produced /ʀ/ in word-initial, intervocalic, and word-final positions next to the vowels /e a o/. Smoothing Spline ANOVAs were used to compare tongue contours within and across phonetic contexts. Differences in the tongue root and tongue body position were also calculated across environments and compared using a measure of coarticulatory resistance. The results revealed that the sound was produced with considerable tongue root retraction and a uvular-pharyngeal tongue body constriction. The tongue root had a high resistance to coarticulatory effects, while the tongue body did not. The results suggest that the tongue root retraction into the pharyngeal cavity results in observed high F1 and low F2 effects associated with unpalatalized rhotic consonants and may explain perceptual similarity between uvular and alveolar rhotics. Articulatory constraints on the tongue root also account for phonotactic distribution of the rhotics across languages.

Acoustics of Kalasha laterals.

Kalasha, a Northwestern Indo-Aryan language spoken in a remote mountainous region of Pakistan, is relatively unusual among languages of the region as it has lateral approximants contrasting in secondary articulation-velarization and palatalization (/ɫ/ vs /lʲ/). Given the paucity of previous phonetic work on the language and some discrepancies between descriptive accounts, the nature of the Kalasha lateral contrast remains poorly understood. This paper presents an analysis of fieldwork recordings with laterals produced by 14 Kalasha speakers in a variety of lexical items and phonetic contexts. Acoustic analysis of formants measured during the lateral closure revealed that the contrast was most clearly distinguished by F2 (as well as by F2-F1 difference), which was considerably higher for /lʲ/ than for /ɫ/. This confirms that the two laterals are primarily distinguished by secondary articulation and not by retroflexion, which is otherwise robustly represented in the language inventory. The laterals showed no positional differences but did show considerable fronting (higher F2) next to front vowels. Some inter-speaker variation was observed in the realization of /ɫ/, which was produced with little or no velarization by older speakers. This is indicative of a change in progress, resulting in an overall enhancement of an otherwise auditorily vulnerable contrast.

Kalasha affricates: An acoustic analysis of place contrasts

AbstractKalasha (Northwestern Indo-Aryan, spoken in Pakistan) exhibits a complex set of ten affricate phonemes, which is exceedingly rare among the world’s languages and not representative of the broader South Asian context. This paper presents results of an acoustic analysis of place contrasts (dental, retroflex, and alveolopalatal) in affricates of four laryngeal specifications (voiceless unaspirated, voiceless aspirated, non-breathy voiced, and breathy voiced). These consonants were produced by four male speakers of Kalasha in a variety of phonetic contexts, resulting in a sample of close to 700 affricate tokens. A series of acoustic analyses of the data revealed that place contrasts in Kalasha affricates are distinguished robustly by both burst/frication spectra and formant transitions, but not by duration, which correlates more with laryngeal features. Place distinctions are somewhat diminished for voiced affricates but are largely unaffected by aspiration and syllable position. Most of these results are consistent with what is known about comparable (yet laryngeally simpler) place contrasts in other languages outside of South Asia. However, some of them are unique and may reflect the typological uniqueness and complexity of Kalasha’s affricate system.

A Review of Voice Disguise in a Forensic Phonetic Context

A Review of Voice Disguise in a Forensic Phonetic Context

Tongue Part Movement Trajectories for /r/ Using Ultrasound.

Because it shows the movement of different parts of the tongue in real time, ultrasound biofeedback therapy is a promising technology for speech research and remediation. One limitation is the difficulty of interpreting real-time ultrasound images of tongue motion. Our image processing system, TonguePART, tracks the tongue surface and allows for the acquisition of quantitative tongue part trajectories. TonguePART automatically identifies the tongue contour based on ultrasound image brightness and tracks motion of the tongue root, dorsum, and blade in real time. We present tongue part trajectory data from 2 children with residual sound errors on /r/ and 2 children with typical speech, focusing on /r/ (International Phonetic Alphabet ɹ) in the phonetic context /ɑr/. We compared the tongue trajectories to magnetic resonance images of sustained vowel /ɑ/ and /r/. Measured trajectories show larger overall displacement and greater differentiation of tongue part movements for children with typical speech during the production of /ɑr/, compared to children with residual speech sound disorders. TonguePART is a fast, reliable method of tracking articulatory movement of tongue parts for syllables such as /ɑr/. It is extensible to other sounds and phonetic contexts. By tracking tongue parts, clinical researchers can investigate lingual coordination. TonguePART is suitable for real-time data collection and biofeedback. Ultrasound biofeedback therapy users may make more progress using simplified biofeedback of tongue movement.

Toward Optimum Quantification of Pathology-Induced Noises: An Investigation of Information Missed by Human Auditory System

Clinical diagnosis of voice disorder and evaluation of therapy outcome heavily rely on accurate quantification of voice quality, which is closely tied to the physiology and function of the laryngeal mechanism. Considering the evaluation methodology of the voice, two main categories of auditory-perceptual assessment and acoustic analysis can be identified. This article presents a new approach for acoustic analysis of voice quality, which brings several advantages to the field. The proposed approach is non-parametric in the sense that it does not require the estimation of the fundamental frequency or spectral response of the vocal tract. This reduces the computational complexity of the measurement and reduces the possible errors due to inaccurate estimation of those parameters. Additionally, the method does not make any assumption about the phonetic context and hence has the potential to be applied to connected speech. The proposed method benefits from the multiresolution structure of the wavelet analysis for estimating the noisy component of a voice in the spectro-temporal domain. The informativeness of the estimated noise for voice quality distinction is examined based on different noise-quantification approaches. It is shown that deviation from the model of the human auditory system (HAS) leads to performance improvement. Through several analyses, it is argued that using models of HAS for quantification of the noise leads to significant loss of information relevant to voice quality. Findings from this article suggest that perception-based measures of voice quality are highly restricted in capturing important aspects of acoustic that could assist with voice quality distinctions. This characteristic is inherent to HAS and cannot be alleviated, highlighting a significant limitation of perception-based measures.

Acoustic Characteristics of Rhotic Vowel Productions of Young Children

Aims: This study examined the acoustic characteristics of perceptually acceptable rhotic vowels produced by young children with and without speech sound disorders (SSDs). Productions were analyzed in relation to the overall rhotic proficiency level of each child, as determined by rhotic vowel and consonant accuracy. The effect of the surrounding phonetic contexts on acoustic realization of rhotic vowels was also examined. Methods: Participants included 18 children aged 2–6 years with and without SSD, grouped by overall rhotic sound proficiency (high rhotic proficiency: ≥70% correct rhotic consonants and vowels; intermediate rhotic proficiency: ≤30% correct rhotic consonants, but ≥70% correct rhotic vowels; low rhotic proficiency: ≤30% correct rhotic consonants and vowels). Target sounds included stressed and unstressed rhotic monophthongs ([ɝ] and [ɚ], respectively) and 4 rhotic diphthongs that differ by pre-rhotic vowel type (/ɪ͡ɚ/, /ɛ͡ɚ/, /ɔ͡ɚ/, /ɑ͡ɚ/). F3 and F3-F2 measures were compared across groups and contexts. Results: No significant differences in F3 and F3-F2 by rhotic sound proficiency group were found in rhotic vowels produced by children with above 70% rhotic vowel accuracy, regardless of their proficiency with rhotic consonants. Acoustic patterns differed by phonetic contexts, but the effect varied by rhotic sound proficiency group. Conclusion: Results showed that once children learn to produce rhotic vowels, they show a comparable degree of rhoticity as those produced by children with high rhotic vowel and consonant accuracy. Results also suggest that rhotic sounds develop earlier in certain phonetic contexts than in others (e.g., [ɝ] before [ɚ]; /ɪ͡ɚ/ and /ɑ͡ɚ/ before /ɔ͡ɚ/).

Competing norms of standard pronunciation. Phonetic analyses on the ‹-ig›-variation in Austria

AbstractThis paper focuses on phonetic variation within the standard German language register of Austria. While the norm status and a high socio-symbolic value are attributed to certain lexical variants of standard language in Austria, the norm and usage status of characteristic phonetic properties remain unclear, due to lack of empirical analyses. By investigating the relation between standard language norms and “standard usage” (Gebrauchsstandard) in Austria, our study aims to close this research gap by using the example of unstressed ‹-ig›. The analyses are based on data gathered from 52 speakers from two generations, covering all Austrian dialect regions. Elicitation settings varied from strongly standardized tasks with a graphic or visual stimulus (reading aloud tasks, picture naming tasks) to translation tasks (translation from dialect into standard) with oral stimuli. The results demonstrate that although ‹-ig› is predominantly pronounced like [ɪk], (socio-)linguistic factors as phonetic context, part of speech, setting, gender and regional background influence the ‹-ig›-variation. In total, the data suggest that German speaking Austrians are situated in a conflict between transnationally diverging norms and intra-nationally varying model speakers of German.

Contextual and speaker variation of glottalic stops in Q'eqchi'

Because of the complex coordination of laryngeal and supralaryngeal articulations, the realization of glottalic stops is highly variable across both languages and speakers. Therefore, it is important to discern the range of variation of glottalic stops. In Q’eqchi’, like other Mayan language, the production of bilabial glottalic stops varies between implosive and ejective. The present study explores whether the phonetic realization of glottalic stops in Q’eqchi’ systematically varies based on phonetic context. We also investigate whether different speakers have preferences for certain realizations of glottalic stops. Examining acoustic measures, we found that (1) both types of glottalic stops (implosive and ejective) are variably produced by speakers, and they are acoustically distinct, (2) ejective-like stops are more likely to occur in word-initial position, and that, (3) individual speakers have preferences towards either producing more implosive-like stops or more ejective-like stops. This study supports the hypothesis that variation of glottalic stops is conditioned by phonetic context, and that differing speakers have different preferences for the realization of these stops.Because of the complex coordination of laryngeal and supralaryngeal articulations, the realization of glottalic stops is highly variable across both languages and speakers. Therefore, it is important to discern the range of variation of glottalic stops. In Q’eqchi’, like other Mayan language, the production of bilabial glottalic stops varies between implosive and ejective. The present study explores whether the phonetic realization of glottalic stops in Q’eqchi’ systematically varies based on phonetic context. We also investigate whether different speakers have preferences for certain realizations of glottalic stops. Examining acoustic measures, we found that (1) both types of glottalic stops (implosive and ejective) are variably produced by speakers, and they are acoustically distinct, (2) ejective-like stops are more likely to occur in word-initial position, and that, (3) individual speakers have preferences towards either producing more implosive-like stops or more ejective-like stops. This study suppo...

Flap articulation and lowered fourth formant

Speakers of North American English are known to use a variety of tap/flap articulations depending on phonetic context (Derrick and Gick, 2011); it is also known that NAE taps/flaps are sometimes associated with a greatly lowered F4 frequency (Warner and Tucker, 2017). It has been less clear whether only certain articulatory variants show this acoustic effect. Since retroflex stops are also associated with lowered F4 (Blumstein and Stevens, 1975), we predict that flap retroflexion is associated with lowered F4. To test this prediction, synchronized ultrasound and audio recordings were made of words containing /t, d/ in a variety of contexts known to give rise to tap/flap variants. Based on visual inspection of ultrasound videos, these were coded as one of four articulatory variants (low tap, high tap, up flap, down flap: Derrick and Gick, 2011); formant frequencies were extracted from the audio at several timepoints relative to the tap/flap. Preliminary results from one speaker support the hypothesis: high taps and down flaps (variants with initial retroflexion) show an F4 drop into the consonant, while high taps and up flaps (variants with final retroflexion) show an F4 rise out of the consonant.

Phonotactics of the Lycian labial glide clusters

Abstract The present article offers a detailed examination of the Lycian phonetic development from a labial glide u̯ {w} into a fricative {b} [v]/[β], which results from contact with an obstruent ([β]/[v] {b} < u̯/C_, AHP: 289). The study of phonetic contexts within each lexeme will allow us to establish new conditions for this change, whether extensions or restrictions, and also to propose some derivations and etymologies.

Normative data on nasalance scores for Farsi speaking children: Influence of age, gender and phonetic context

BackgroundNormative nasalance scores are essential for treatment and assessment purposes for clinicians. The purpose of the current study is to determine the normative data on nasalance scores for Farsi speaking children. Comparing to the obtained normative data, hypernasal or hyponasal speech (i.e. the higher or lower nasalance score) can be evaluated. MethodsNinety-six Farsi speakers ranging from 4 years to 11 years old with typically developing speech were included in this study. Mean nasalance scores were obtained using the Universal Parameters System sentences with nasometer II (model 6450). The nasalance data was also analyzed for any gender, age and phonetic context dependence. ResultsNormative nasalance values were determined, Nno differences were found for gender and age for the school-aged children. With an increasing in age, the nasalance score also increased in oral passages but not in nasal passages. Furthermore, amongst the nasalance scores there was a significant difference between these four categories of sentences: 7 sentences loaded with oral stop consonants, 5 sentences loaded with oral fricative consonants, 2 sentences loaded with oral affricatives and 3 sentences loaded with nasal consonants. ConclusionsThe absence of age and gender differences led to a computation of an overall mean nasalance score for different types of phonetic context (Nasal, oral and oronasal). These normative nasalance data for Farsi speaking children will provide important reference information during speech assessments of children with velopharyngeal insufficiency.

Noninvasive neurostimulation of left temporal lobe disrupts rapid talker adaptation in speech processing

Talker adaptation improves speech processing efficiency by reducing possible mappings between talkers’ speech acoustics and listeners’ phonemic representations. We investigated the functional neuroanatomy of talker adaptation by applying noninvasive neurostimulation (high-definition transcranial direct current stimulation; HD-tDCS) to left superior temporal lobe while participants performed an auditory word identification task. We factorially manipulated talker variability (single vs. mixed talkers) and speech context (isolated words vs. connected speech), measuring listeners’ speech processing efficiency under anodal, cathodal, or sham stimulation. Speech processing was faster for single talkers than mixed talkers, and connected speech reduced the additional processing costs associated with mixed-talker speech. However, the beneficial effect of connected speech in the mixed-talker condition was significantly attenuated under both anodal and cathodal stimulation versus sham. Stimulation of left superior temporal lobe disrupts the brain’s ability to use local phonetic context to rapidly adapt to a talker, revealing this region’s causal role in talker adaptation.

Duration as a Phonetic Cue in Native and Non‑Native American English

This vowel study looks at the intricate relationship between spectral characteristics and vowel duration in the context of American English vowels, both from a native speaker (NS) and non-native speaker (NNS) perspective. The non-native speaker cohort is homogeneous in the sense that all speakers have Serbian as their mother tongue, but have been long-time residents of the US. The phonetic context investigated in this study is /bVt/, where V is one of the American English monophthongs /i ɪ u ʊ ε æ ʌ ɔ ɑ/. The results of the acoustic analysis show that the NNS vowels are generally longer than the NS vowels. Furthermore, NNSs neutralise the vowel quality of two tense and lax pairs of vowels, /i ɪ/ and /u ʊ/, and rely more heavily on the phonetic duration when prononuncing them.

Gradual conventionalization of pragmatic inferences

Abstract The alternation in Spanish between y and e on the one hand, and u and o in the other, is examined. It is proposed that the standard account under which the choice of one variant over the other is sensitive only to the phonetic context is incomplete. Specifically, the paper argues that pragmatic inferences that typically appear cross-linguistically associated with these connectors, and that result in asymmetric interpretations, are not favoured in Spanish with the morphological variants e and u, which favour symmetric interpretations. The paper proposes that the relevant pragmatic inferences have been partially conventionalized for y and o, but that this conventionalization has not occurred in the case of e and u for the reason that they are much less frequently used. Thus, discussion and data offer a view of a stage in a gradual process of semantic change via conventionalization of pragmatic inferences.

Rhotic vowel accuracy and error patterns in young children with and without Speech Sound Disorders

Rhotic vowels are known to be difficult sounds for children to learn, but their acquisition has not been investigated in detail. This study examined rhotic vowel development, including accuracy and error patterns, in young children across age, diagnostic groups, and phonetic contexts. Participants included 34 children from 2 to 6 years of age with and without speech sound disorders (SSD). Data included 36 elicited single words containing stressed and unstressed rhotic monophthongs ([ɜ˞] and [ə˞]) and four rhotic diphthongs (/ɪ͡ə˞/, /ε͡ə˞/, /ɔ͡ə˞/, and /ɑ͡ə˞/). Additional words containing non-rhotic vowels and the consonant /ɹ/ were also included for comparison. A significant difference in rhotic vowel production accuracy was found between diagnostic groups (with vs. without SSD), but not across age groups or phonetic contexts. Yet, the accuracy scores for children without SSD were bimodally distributed, showing overlapping patterns with children with SSD. Error patterns were examined for all children with low rhotic vowel accuracy scores, regardless of age and diagnostic status. Results of this study highlight the variability in rhotic vowel development in young children and the importance of studying rhotic sounds using overall rhotic sound accuracy of each individual child regardless of their age or diagnostic status. Clinically, systematic within-speaker error patterns suggest the need for the detailed error pattern assessment of rhotic sounds.

Backwards and indirect context effects in accommodating gender differences in speech

Listeners accommodate large amount of acoustic variability across talkers and phonetic contexts when categorizing speech sounds. This study examines accommodation of gender-related talker differences, which can occur in reverse (in sound sequence AB, sound B affects perception of sound A), suggesting more complicated mechanisms than peripheral auditory contrast enhancement alone. A continuum of fricatives “sh” and “s” was preappended to vowels whose acoustic parameters were manipulated by various properties ranging from typically feminine to masculine and vice versa. Other conditions provided matching or mis-matching visual cues. We examined the weighting of cues to talker gender, the duration of exposure to sound needed to obtain the full context effect, and the influence of hearing status, using listeners with cochlear implants (CIs). CI listeners used different cues to show a comparable gender context effect and also appear to use proxy cues for acoustic properties perceived more clearly by people with normal hearing (NH). All listeners showed implicit knowledge of how acoustic patterns are selectively reverse-compatible with different articulation sequences. NH listeners needed only 30 ms for full adjustment, while CI listeners’ data were much more variable. Collectively, these results show a complex process of auditory, visual and gesture-aware adjustment to a talker’s voice.Listeners accommodate large amount of acoustic variability across talkers and phonetic contexts when categorizing speech sounds. This study examines accommodation of gender-related talker differences, which can occur in reverse (in sound sequence AB, sound B affects perception of sound A), suggesting more complicated mechanisms than peripheral auditory contrast enhancement alone. A continuum of fricatives “sh” and “s” was preappended to vowels whose acoustic parameters were manipulated by various properties ranging from typically feminine to masculine and vice versa. Other conditions provided matching or mis-matching visual cues. We examined the weighting of cues to talker gender, the duration of exposure to sound needed to obtain the full context effect, and the influence of hearing status, using listeners with cochlear implants (CIs). CI listeners used different cues to show a comparable gender context effect and also appear to use proxy cues for acoustic properties perceived more clearly by people with...

Variations of F2 range measures in mild dysarthria and healthy talkers

Diminished vocalic segments indexed by the extent, duration, and rate of change in the second formant (F2) are common in dysarthria and are known to reduce intelligibility in adults with multiple sclerosis (MS; Hartelius et al., 2010). When F2 range was measured over select sentences from a reading passage, results indicated that the F2 range may not be sensitive to mild dysarthria in MS (Rosen et al., 2008). However, measures of the F2 range should be further explored for sentences imposing greater speech-motor and cognitive-linguistic demands. As such, phonetic contexts requiring even small vocal tract movement may contribute to intelligibility for speakers with mild dysarthria in MS. Thus, this study will investigate the impact of sentence length and complexity on within-speaker variations of F2 movement and group differences for 12 adults with mild dysarthria in MS and 12 healthy talkers. Various F2 range measures shown to be sensitive to speech motor capabilities will be obtained from 11 audio-recorded sentences varying in length and complexity. The results of this study will provide a better understanding of the speech motor control capabilities of speakers with mild dysarthria with relatively preserved intelligibility in MS. [Work supported by ASH Foundation and the University at Buffalo MDRF.]Diminished vocalic segments indexed by the extent, duration, and rate of change in the second formant (F2) are common in dysarthria and are known to reduce intelligibility in adults with multiple sclerosis (MS; Hartelius et al., 2010). When F2 range was measured over select sentences from a reading passage, results indicated that the F2 range may not be sensitive to mild dysarthria in MS (Rosen et al., 2008). However, measures of the F2 range should be further explored for sentences imposing greater speech-motor and cognitive-linguistic demands. As such, phonetic contexts requiring even small vocal tract movement may contribute to intelligibility for speakers with mild dysarthria in MS. Thus, this study will investigate the impact of sentence length and complexity on within-speaker variations of F2 movement and group differences for 12 adults with mild dysarthria in MS and 12 healthy talkers. Various F2 range measures shown to be sensitive to speech motor capabilities will be obtained from 11 audio-record...