Acoustic and articulatory correlates of contrastive focus

Abstract

This paper aims at examining production of French speakers in marking focus-induced prominence at the articulatory and acoustic levels. The corpus consisted of CVC syllables, where C corresponds to one of the stops /p t k/ and V is one of the vowels /i a u/. Target words were embedded in carrier sentences elicited in two prosodic conditions: neutral (unfocused) and under contrastive focus. Four adult speakers (all native speakers of French) pronounced ten repetitions of each sequence. The audio signal and tongue shapes were recorded using a digital camera and a SONOSITE 180 ultrasound. Formant frequencies, rms values, duration, and tongue contours corresponding to each vowel were extracted. Analyses show that prosodic context has a significant effect on acoustic and articulatory data for all vowels, increasing F1 and F2 under contrastive focus, compared to the neutral context. However, this stable acoustic pattern is achieved by various articulatory strategies across subjects. For example, 2 subjects produced /u/ under focus with a lower tongue body than in the neutral context, whereas the remaining 2 subjects had a higher tongue body in focused syllables. Results are compared to previous studies on articulatory and acoustic correlates of prosodic structure in French and English.