Hybrid Acoustic-Lexical Deep Learning Approach for Deception Detection

@inproceedings{Mendels2017HybridAD,
  title={Hybrid Acoustic-Lexical Deep Learning Approach for Deception Detection},
  author={Gideon Mendels and Sarah Ita Levitan and Kai-Zhan Lee and Julia Hirschberg},
  booktitle={INTERSPEECH},
  year={2017}
}
Automatic deception detection is an important problem with far-reaching implications for many disciplines. We present a series of experiments aimed at automatically detecting deception from speech. We use the Columbia X-Cultural Deception (CXD) Corpus, a large-scale corpus of within-subject deceptive and non-deceptive speech, for training and evaluating our models. We compare the use of spectral, acoustic-prosodic, and lexical feature sets, using different machine learning models. Finally, we… Expand

Figures, Tables, and Topics from this paper

Deception Detection and Analysis in Spoken Dialogues based on FastText
TLDR
A deception detector for dialogue utterances is constructed by giving labels of deception to FastText and the resultant detector achieved significantly higher accuracy than deception-detection by humans. Expand
JOINT LEARNING OF CONVERSATIONAL TEMPORAL DYNAMICS AND ACOUSTIC FEATURES FOR SPEECH DECEPTION DETECTION IN DIALOG GAMES
TLDR
A proposed joint learning framework to detect deception by simultaneously considering variations and patterns of the conversation using both interlocutor's acoustic features and their conversational temporal dynamics achieves an unweighted average recall (UAR) of 74.71%. Expand
Acoustic-Prosodic Indicators of Deception and Trust in Interview Dialogues
TLDR
The results show that acoustic-prosodic features are highly effective at classifying deceptive speech, and have implications for automatic deception detection and the development of synthesized speech that is trustworthy. Expand
A Study of Data Scarcity Problem for Automatic Detection of Deceptive Speech Utterances∗
In recent years there have been a lot of interest in the task of contactless automatic detection of deception in speech utterances. This task belongs to the paralinguistic area that studies theExpand
Comparative Analysis of Classification Methods for Automatic Deception Detection in Speech
TLDR
A short survey of databases that contain both deceptive and truthful speech samples, as well as recently developed deception detection systems that were proposed within the framework of computational paralinguistic challenge ComParE-2016 and other scopes are presented. Expand
Interpretable Multimodal Deception Detection in Videos
TLDR
This manuscript provides a principled way to capture rich information into a coherent model and proposes an end-to-end framework DEV to detect DEceptive Videos automatically and results on real-world videos demonstrate the effectiveness of the proposed framework. Expand
Automated Recognition of Sincere Apologies from Acoustics of Speech
TLDR
A multi-model fusion framework to identify sincerely delivered apologies by modelling difference between acoustics of sincere and insincere utterances is proposed and the efficacy of this framework is benchmarked using the Sincere Apology Corpus. Expand
Deep Personality Recognition for Deception Detection
TLDR
It is shown that including personality recognition improves the performance of deception detection on the Columbia X-Cultural Deception corpus by more than 6% relative, achieving new state-of-the-art results on classification of phrase-like units in this corpus. Expand
Applying Ensemble Learning Techniques and Neural Networks to Deceptive and Truthful Information Detection Task in the Flow of Speech
TLDR
This paper presents the results of experiments on applying ensemble learning techniques and neural networks to a paralinguistic analysis of deceptive and truthful statements in the flow of speech and proposes using a mixture of such methods. Expand
Deception in the eyes of deceiver: A computer vision and machine learning based automated deception detection
TLDR
For the first time, fine-grained level eyes and facial micro-movements are investigated to identify the distinctive features that provide significant clues for the automated deception detection. Expand
...
1
2
3
4
...

References

SHOWING 1-10 OF 23 REFERENCES
Distinguishing deceptive from non-deceptive speech
TLDR
Results from a study seeking to distinguish deceptive from non- deceptive speech using machine learning techniques on features extracted from a large corpus of deceptive and non-deceptive speech are presented. Expand
Verbal and Nonverbal Clues for Real-life Deception Detection
TLDR
This paper develops a new deception dataset consisting of videos from reallife scenarios, and builds deception tools relying on verbal and nonverbal features, and shows that these results outperform the human capability of identifying deceit. Expand
Cross-Cultural Production and Detection of Deception from Speech
Detecting deception from different dimensions of human behavior has been a major goal of research in psychology and computational linguistics for some years and is currently of considerable interestExpand
Is Deception Emotional? An Emotion-Driven Predictive Approach
TLDR
A method for automatically detecting deceptive speech by relying on predicted scores derived from emotion dimensions such as arousal, valence, regulation, and emotion categories, derived from task-dependent models trained on the GEMEP emotional speech database is proposed. Expand
Adieu features? End-to-end speech emotion recognition using a deep convolutional recurrent network
TLDR
This paper proposes a solution to the problem of `context-aware' emotional relevant feature extraction, by combining Convolutional Neural Networks (CNNs) with LSTM networks, in order to automatically learn the best representation of the speech signal directly from the raw time representation. Expand
Verification and Implementation of Language-Based Deception Indicators in Civil and Criminal Narratives
Our goal is to use natural language processing to identify deceptive and non-deceptive passages in transcribed narratives. We begin by motivating an analysis of language-based deception that reliesExpand
Speech recognition with deep recurrent neural networks
TLDR
This paper investigates deep recurrent neural networks, which combine the multiple levels of representation that have proved so effective in deep networks with the flexible use of long range context that empowers RNNs. Expand
Lying Words: Predicting Deception from Linguistic Styles
TLDR
The current project investigated the features of linguistic style that distinguish between true and false stories, and found that liars showed lower cognitive complexity, used fewer self-references and other- References, and used more negative emotion words than truth-tellers. Expand
Dropout: a simple way to prevent neural networks from overfitting
TLDR
It is shown that dropout improves the performance of neural networks on supervised learning tasks in vision, speech recognition, document classification and computational biology, obtaining state-of-the-art results on many benchmark data sets. Expand
Finding Deceptive Opinion Spam by Any Stretch of the Imagination
TLDR
This work develops and compares three approaches to detecting deceptive opinion spam, and develops a classifier that is nearly 90% accurate on the authors' gold-standard opinion spam dataset, and reveals a relationship between deceptive opinions and imaginative writing. Expand
...
1
2
3
...