Retour vers liste

Détail de la contribution

Auteur: Volker DELLWO

Co-Auteur(s): Agnes KOLMER, University of Zurich, Switzerland Daniel FRIEDRICHS, University of Zurich, Switzerland

Perception of French accentuation by Swiss German and Standard German speakers

Abstract/Résumé: Speakers of Swiss German (SwG) typically perceive French words with a stress on the first syllable while Standard German (StG) speakers perceive the same words with final stress. In the present research we studied whether this difference is based on (a) a varying interpretation of acoustic cues by SwG and StG listeners (Qian, 2008) or (b) different expectations by the two listener groups towards French word accent (Eriksson, 2002). In two experiments we tested the perception of French word stress in tri-syllabic words by SwG, StG and French listeners using stimuli in which the accentuation changed from initial syllable to syllable stress. Listeners either knew (experiment I) or did not know (experiment II) that the acoustic information they were interpreting was French. Stress shifts were produced by changing pitch, duration and amplitude on a seven-step acoustic continuum from a initial to a final syllable stress. In Experiment I listeners were tested on natural speech to which the accent changes were applied synthetically (real-word condition). In Experiment II, different listeners were tested on delexicalized speech in which all vowels were turned into /a/- and all consonants into /s/-sounds (Ramus and Mehler, 1999) while the acoustic prosodic information was identical to the one in the real-word stimuli in experiment I (sasasa-condition). Results from 5 listeners of each StG and SwG showed that the two listener groups only deviate in their perception in the real-word condition, suggesting that the interpretation of acoustic prosodic information between the two listener groups is identical without knowledge of the language (results from larger listener groups as well as a French group are currently being evaluated). Results suggest that language specific expectations about word stress systems have a large influence on how listeners perceive them. References: - Eriksson, A., E. Grabe und H. Traunmüller (2002) Perception of syllable prominence by listeners with and without competence in the tested language. In: Proceedings of Speech Prosody 2002, 275-278. - Qian, W. (2008) L2 Stress Perception: The reliance on different acoustic cues. In: Electronic Proceedings of Speech Prosody 2008. - Ramus, F. and Mehler. J. (1999) Language identification with suprasegmental cues: A study based on speech resynthesis. In: J. Acoust. Soc. Am. 105(1), 512-521.