A prosodic cue that invites back-channel responses in Arabic

Yaffa Al Bayyari, University of Texas at El Paso

Abstract

To be a good listener requires active listening, which can be achieved partly by producing back-channel feedback. Back-channels are short utterances produced by the listener while the speaker is having the floor; typical back-channels in English are uh-huh and ok. Second-language learners, even if masters of grammar and vocabulary, can easily appear uninterested when they do not show responsiveness in real face-to-face conversations. In Arabic, there is a lack of resources that teach L2 learners when to produce back-channels in dialog. Work in other languages suggests that there are times where back-channels are especially welcomed, and these are mostly indicated by how the speaker is saying a phrase—the prosody of the utterances, in other words. Using the CALLHOME Egyptian Arabic corpus and the Iraqi Arabic corpus (Nigel & Novick & Salamah, 2005), a corpus-based analysis has been conducted and this resulted in finding the most frequent prosodic feature used as a cue for back-channel feedback in Arabic to be a steep pitch downslope. The performance of this feature as a predictive rule was 62% coverage and 17% accuracy. The next most frequent cue (accounting for 14% of the total back-channels) is an upturn in pitch, whose predictive rule was 19% coverage and 15% accuracy on an Arabic corpus. Further work was done on the role of gestures in cueing back-channel responses, using a video-recorded Arabic corpus, showing that gestures do tend to co-occur with subsequent back-channels. Moreover, experiments were conducted to find how the pitch downslope cue is perceived by non-Arabic speakers. The results showed that pitch downslope is not perceived as a cue for back-channel feedback and was in fact perceived negatively by American-English speakers.

Subject Area

Computer science

Recommended Citation

Al Bayyari, Yaffa, "A prosodic cue that invites back-channel responses in Arabic" (2007). ETD Collection for University of Texas, El Paso. AAI1448752.
https://scholarworks.utep.edu/dissertations/AAI1448752

Share

COinS