Combining Acoustic-Prosodic, Lexical, and Phonotactic Features for Automatic Deception Detection

Sarah Ita Levitan, Guozhen An, Min Ma, Rivka Levitan, Andrew Rosenberg, Julia Hirschberg


Improving methods of automatic deception detection is an important goal of many researchers from a variety of disciplines, including psychology, computational linguistics, and criminology. We present a system to automatically identify deceptive utterances using acoustic-prosodic, lexical, syntactic, and phonotactic features. We train and test our system on the Interspeech 2016 ComParE challenge corpus, and find that our combined features result in performance well above the challenge baseline on the development data. We also perform feature ranking experiments to evaluate the usefulness of each of our feature sets. Finally, we conduct a cross-corpus evaluation by training on another deception corpus and testing on the ComParE corpus.


DOI: 10.21437/Interspeech.2016-1519

Cite as

Levitan, S.I., An, G., Ma, M., Levitan, R., Rosenberg, A., Hirschberg, J. (2016) Combining Acoustic-Prosodic, Lexical, and Phonotactic Features for Automatic Deception Detection. Proc. Interspeech 2016, 2006-2010.

Bibtex
@inproceedings{Levitan+2016,
author={Sarah Ita Levitan and Guozhen An and Min Ma and Rivka Levitan and Andrew Rosenberg and Julia Hirschberg},
title={Combining Acoustic-Prosodic, Lexical, and Phonotactic Features for Automatic Deception Detection},
year=2016,
booktitle={Interspeech 2016},
doi={10.21437/Interspeech.2016-1519},
url={http://dx.doi.org/10.21437/Interspeech.2016-1519},
pages={2006--2010}
}