This study explores the articulatory characteristics of plain and palatalized fricatives in Romanian. Based on earlier acoustic findings, we hypothesize that there are differences in tongue raising and fronting depending on the primary place of articulation, with more subtle gestures produced in the vicinity of the palatal area. We also predict more individual variation in the realization of secondary palatalization in postalveolars, based on general cross-linguistic patterns. Ten native speakers participated in an ultrasound experiment. The stimuli included real words containing labial, dental, and postalveolar fricatives. The fricatives at all three places were either plain or palatalized word-finally (the only position available for secondary palatalization in this language). Tongue contours at the consonant midpoint were compared using Smoothing Spline ANOVAs individually with radius distance from the ultrasound probe. The findings indicate differences in tongue shape between plain and palatalized consonants, with stronger palatalization effects in labials compared to coronals, as well as in dentals compared to postalveolars. The latter also revealed higher individual variation. Our findings thus suggest that tongue configurations for secondary palatalization in Romanian differ by place of articulation. The contrast is also overall less robust in postalveolars, confirming previous reports and explaining its rarity cross-linguistically.

This study extends the use of ultrasound methodology to stops in Eastern Oromo (Cushitic; Ethiopia) to examine the link between gemination, laryngeal features, and tongue shape.Ultrasound data were collected from 5 native speakers of Eastern Oromo. Tokens consisted of 12 repetitions per speaker of [t h , t, d, â] and six of [tt h , tt, dd, ââ] in the environment of a a. Tongue images at the point of maximum constriction during the stop closure were traced following [1] and their coordinates submitted to linear mixed effects models.Results indicated differences in tongue shape between singletons and geminates, especially for ejectives and implosives. Singleton ejectives displayed raised tongue bodies not found in geminate ejectives. Singleton implosives resembled voiceless stops, but geminate implosives were variably produced with tongue body raising.I suggest that the results can be attributed to fortition in geminates. Tongue body raising in singleton ejectives may be an enhancement strategy to the ejective contrast that is not necessary in longer geminates. The singleton implosive resembling a voiceless aspirated stop is predicted by [15] while the geminate tongue body raising may be retraction, c.f. [2]. The results support a link between tongue, larynx, and gemination.

This paper investigates laryngeal contrasts in first (L1) and second (L2) language speakers of Hul’q’umi’num’, a dialect of Halkomelem (Salish). The language, which lacks a thorough acoustic description of its consonants, is highly endangered with 50-100 native speakers, but has a growing number of L2 speakers. Since the learners come from an English background and are therefore unfamiliar with glottalized consonants, the sounds can prove challenging. This study addresses this issue by examining how L1 and L2 speakers pronounce plain and ejective stops in terms of what acoustic correlates they’re using. Tokens of words read in isolation will be analyzed from L1 and L2 speakers participating in a language course in Duncan, BC. Tokens of both speaker groups will be classified by phoneme and acoustic measures of duration (e.g., voice onset time, closure duration), centre of gravity, and vowel coarticulation (e.g., spectral tilt, F0, rise time) will be made. The results of the two groups will be compared with statistical analysis, and the findings will be used to create guidelines to assess future learners’ and teach pronunciation. The findings will also contribute to a broader understanding of how Hul’q’umi’num’ consonants fit into voicing and ejective typology.

Dene or North Slavey is a Dene (Athabaskan) language with nine affricates: /ts, tsʰ, ts’, t∫, t∫ʰ, t∫’, tɬ, tɬʰ, and tɬ’/. This paper provides a first investigation into the acoustics of these sounds using ~2000 tokens from eight speakers from Déline, NT, Canada. Within the lateral series, [tɬ] was found to be realized in a number of ways including [tɬ], [tɬl], [tl], and [tɬɮ], with phonetic voicing to varying degrees in the release. In contrast, [tɬʰ] was generally produced with a brief period of aspiration following the frication portion of the affricate, while ejectives were often followed by a brief period of silence between the frication and the vowel onset. COG measurements of [tɬ] (from the voiceless portion of the release) were lower than the alveolar and post-alveolar affricates, averaging 3500 Hz (s.d. = 2175 Hz). However, COG variation across and within speakers and words suggests varying place of articulation. Place of articulation may be less crucial to the identification of these sounds than manner cues such as the lateral component. Further work is underway to investigate differences in the distribution of energy in the frication and F0 perturbations between the different phonation types, and to investigate the alveolar and post-alveolar affricates in more depth.

The nature of edge intonational contours as well as acoustics of fricatives have generally been independently discussed in the literature (Hughes & Halle 1956; Ladd 1996; Gussenhoven 2004 inter alia). Voiceless consonants were traditionally conceived as irrelevant to the study of utterance-level intonation and thought merely to interrupt pitch contours (Bolinger 1964). However, Niebuhr (2012) proposes that the two domains interact, reporting that German fricatives exhibit relatively higher centre of gravity (CoG) and higher acoustic energy in the context of rising intonation. This phenomenon, known as segmental intonation, has been found in some languages (Polish, Zygis et al. 2014; Dutch, Heeren 2015), but remains controversial in others (English, Niebuhr p.c.). We test this hypothesis by replicating the reading task in Niebuhr’s (2012) study for English and also extending to a tonal language, Cantonese, in which F0 is used grammatically to distinguish words, in addition to intonation. Preliminary results from 10 speakers suggest that segmental intonation in the form of higher CoG and intensity exists in English, but not in Cantonese. With additional data recently collected, we hope to confirm these findings and contribute to determining whether the segmental intonation is an epiphenomenon of speech production in general or not.