Articulatory trade-offs between voicing and vowel height in (post)alveolar fricatives. A multi-level phonetic study.

The aim of our study is to investigate the supraglottal correlates of voicing in Italian alveolar and postalveolar fricatives in low and high vowel contexts. We address the general issues of how phonetic granularity enhances paradigmatic contrasts in language-specific ways (e.g. [2], [9]). Voicing and frication have conflicting aerodynamic requirements: a transglottal pressure difference is required for voicing, while a pressure drop across the fricative oral constriction is necessary for frication. Several articulatory strategies can be used to overcome the biomechanical constraints of voicing during frication. Voiced sibilants such as /z/ and /ʒ/ are generally shorter, more fronted, and articulated with posterior cavity expansion and a smaller groove compared to /s/, /ʃ/ in German, English, Croatian, French, Turkish (e.g. [4], [5], [6], [8], [10]). There is cross-subject and cross-language variation in the way the different strategies are implemented, though. Moreover, (post)alveolar fricatives are reported to be longer when the surrounding vowels are high, compared to when the surrounding vowels are low (e.g. [7], [11]), which has been interpreted as anticipatory scanning (shorter distance to be covered by the articulators from the fricative constriction to the vowel); however, only voiceless fricatives have been investigated. It has also been reported that the cricothyroid muscle is more active in high vowels, favouring the periodicity of vibration (e.g. [12], [3]). The current study is a further step of evidence addressing in particular the question of how vowel height interacts with supralaryngeal voicing adjustments. Alternative hypotheses will be tested: 1) voiced fricatives in a high vowel context are longer; given the difficulty to maintain voicing for a longer time, articulatory strategies for voicing are more prominent in high vowel than in low vowel environments; 2) voiced fricatives in a high vowel context are longer; since high vowels support voicing, there are smaller articulatory differences between voiced and voiceless fricatives in this context; 3) an earlier consonantal release is found in high vowel contexts as the main strategy to maintain voicing, independent on vowel quality. Four female speakers of a Tuscan variety of Italian aged 31-39 were recorded via SynchroLing while producing 5 repetitions of nonword stimuli containing /s sː z ʃ ʃː ʒ/. SynchroLing is a platform for multilevel phonetic data acquisition and analysis which allows real-time synchronization of audio, UTI and EPG channels ([1]). Nonwords were trisyllables with stress on the antepenult and consonants were symmetrically flanked by /a/ or /i/ (e.g. /paˈsapo/, /piˈsipo/). Mixed effects models were used to determine the role of four predictors (i.e. voicing, gemination [for voiceless fricatives only], vowel height and place of articulation) on consonant duration. In order to compare voiceless and voiced fricatives, confidence intervals of contact anteriority (CA), contact posteriority (CP), groove extension or contact centrality (CC) and center of gravity (CoG) were calculated. Tongue splines are analysed by means of generalized additive models (GAM). The analysis is still in progress and the preliminary results show that voiced fricatives are significantly longer in the /i/ context, however the shortening effect of voicing is stronger in the /i/ than in the /a/ environment (Figure 1). Articulatory differences depend on the place of articulation of the fricatives as well as on the vocalic context, although most differences neutralize in the /i/ context (Figure 2). Cross-subject variation is also found. We discuss the findings to show how secondary phonetic cue variation enhance voicing differences in language-specific ways. [1] Chen, C., Celata, C. and Ricci, I. 2017. An EPG + UTI study of syllable onset and coda coordination and coarticulation in Italian. In Bertini, C., Celata, C., Lenoci, G., Meluzzi, C. and Ricci, I. (Eds.), Social and biological factors in speech variation. Studi AISV 3. Milano, Officinaventuno. [2] Cho, T. 2015 Language Effects on Timing at the Segmental and Suprasegmental Levels. In Redford, M.A. (Ed.), Handbook of Speech Production. John Wiley, 505-529. [3] Ishikawa, C.C., Pinheiro, T.G., Hachiya, A., Montagnoli, A.N., Tsuji, D.H. 2017. Impact of Cricothyroid Muscle Contraction on Vocal Fold Vibration: Experimental Study with High-Speed Videoendoscopy. Journal of Voice 31(3): 300-306. [4] Jongman, A, Wayland R. and Wong, S. 2000. Acoustic characteristics of English fricatives. Journal of the Acoustical Society of America 108 (3): 1252-1263. [5] Liker, M. and Gibbon, F. 2013. Differences in EPG contact dynamics between voiced and voiceless lingual fricatives. Journal of the International Phonetic Association 43 (1). [6] Proctor, M.I., Shadle, C. and Iskharous, K. 2010. Pharyngeal articulation in the production of voiced and voiceless fricatives. Journal of the Acoustical Society of America 127(3): 1507-1518. [7] Schwartz, M.F. 1969. Influence of vowel environment upon the duration of /s/ and /∫/. Journal of the Acoustical Society of America, 46 (2B): 480-481. [8] Solé, M.-J. 2018. Articulatory adjustments in initial voiced stops in Spanish, French and English. Journal of Phonetics 66: 217-241. [9] Solé, M.J. and Ohala, J.J. 2010. What is and what is not under the control of the speaker. Intrinsic vowel duration. In Fougeron C., Kühnert B., D’Imperio M. & Vallée, N. (Eds.) Papers in Laboratory Phonology 10. Berlin: de Gruyter, 607-655. [10] Ünal-Logacev, Ö., Fuchs, S. and L. Lancia (2018) A multimodal approach to the voicing contrast in Turkish: Evidence from simultaneous measures of acoustics, intraoral pressure and tongue palatal contacts. Journal of Phonetics 71: 395-408. [11] Weglarski, A., Sewall, A., Schiavetti, N. and Metz, D.E. 2000. Effect of vowel environment on consonant duration: an extension of normative data to adult contextual speech. Journal of Communication Disorders 33(1): 1-10. [12] Whalen, D., Gick, B., Kumanda, M. and Honda, K. 1999. Cricothyroid activity in high and low vowels: exploring the automaticity of intrinsic F0. Journal of Phonetics 27: 125-142.