Date of Degree


Document Type


Degree Name





Sarah Ita Levitan

Subject Categories

Computational Linguistics | Discourse and Text Linguistics


speech processing, entrainment, trust


This thesis explores the possibility of using features of speech as possible predictors of perceived trust. It specifically focuses on entrainment, the tendency of participants in a conversation to unconsciously adapt their manner of talking to become more or less similar to each other. We compile a corpus of interviews conducted in English, with different levels of formality and discussing different topics. With it, we distribute a set of surveys to assess raters’ judgment of the participants, focusing on whether they believe the interviewee is trusted by their conversational partner, and whether they find the interlocutors themselves trustworthy. We also extract a set of speech features from evenly distributed, but randomly selected, conversational turn triplets from each interview, and we compute global and local proximity, convergence and synchrony measurements with said features. Then, we measure the strength and significance of the correlations between these measurements and the trust judgments obtained from the surveys. We find that local proximity in pitch and intensity, and sometimes in harmonics-to-noise ratio and speaker rate, are potentially reliable predictors of whether listeners believe the speakers trust each other. Local proximity in the same features, along with jitter and shimmer, were found to be good predictors of whether listeners themselves will find the speakers trustworthy. These correlations can be direct or indirect depending on whether we look at a single pair of turns or multiple. This suggests that a variance in proximity as the conversation progresses, which indicates greater engagement, could be seen as more trustworthy than a constant similarity.

This work is embargoed and will be available for download on Monday, September 30, 2024

Graduate Center users:
To read this work, log in to your GC ILL account and place a thesis request.

Non-GC Users:
See the GC’s lending policies to learn more.