Applying Ensemble Learning Techniques and Neural Networks to Deceptive and Truthful Information Detection Task in the Flow of Speech

Velichko, Alena; Budkov, Victor; Kagirov, Ildar; Karpov, Alexey

doi:10.1007/978-3-030-32258-8_56

Cited by 3 publications

(7 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For comparison, single models of Cabtoost, XGBoost and LightGBM have achieved results in terms of F-score of 84.1%, 84.6%, and 85.0% respectively. The achieved empirical results are highly competitive and comparable with the results presented by other researchers [11,12,17,18,21,20,23] (see Table 1). We present our results achieved with the use of two corpora and the approach described above.…”

Section: Discussion Of the Resultssupporting

confidence: 86%

“…Approach from [11] F-score = 63.9%, Precision = 76.1% Approach from [12] UAR = 74.9% Baseline system [17] UAR = 68.3% Approach from [18] Accuracy (max) = 75.0% Approach from [21] UAR (max) = 70.0% Approach from [20] UAR = 73.5%, F-score = 75.0%, Precision = 77.0% Approach from [23] Accuracy…”

Section: Approach Classification Resultsmentioning

confidence: 99%

“…This system was based on Support Vector Machines (SVM) and achieved results in term of UAR = 73.5%, mean F-score = 75.0% and Precision = 77.0%. In [21] the implementation of ensemble methods and neural networks were proposed. The ensemble consists of k-Nearest Neighbours, Random Forest and Neural Network have achieved the UAR of 65.0% and 70.0%, in case of average voting and majority voting correspondingly.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automatic Detection of Deceptive and Truthful Paralinguistic Information in Speech using Two-Level Machine Learning Model

Velichko¹,

Saint-Petersburg²,

Karpov³

2021

Computational Linguistics and Intellectual Technologies

View full text Add to dashboard Cite

In this work, we present a novel approach to one of computational paralinguistic tasks -automatic detection of deceptive and truthful information in human's speech. This task belongs to the aspects of destructive behaviour and was first presented at the International INTERSPEECH Computational Paralinguistics Challenge ComParE in 2016. The need of contactless method for deception detection follows from the fact that existing contact-based approaches such as polygraphs and lie detectors have multiple restrictions, which significantly limit their usage. Both for training and testing of the proposed models we used two English-language corpora (Deceptive Speech Database and Real-Life Trial Deception Detection Dataset). We extracted tree sets of acoustic features from those audio samples using openSMILE toolkit. The proposed approach includes preprocessing of the extracted acoustic features with the usage of methods for data augmentation and dimensionality reduction of feature space. We have got 1680 speech utterances and 986-dimensional informative feature vector for each utterance. The main part of the proposed approach is two-level recognition model, where the first level includes three models of gradient boosting (Catboost, XGBoost and LightGBM). The second level consists of logistic regression-based model for final prediction on truthfulness or deceptiveness that takes into account predictions from the first level. Using this approach, we have achieved the result of classification in terms of F-score = 85.6%. The proposed approach can be used both independently and as a component of multimodal systems for detection of deceptive and truthful utterances in speech, as well as in systems for detection of a destructive behaviour.

show abstract

Section: Discussion Of the Resultssupporting

confidence: 86%

Section: Approach Classification Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automatic Detection of Deceptive and Truthful Paralinguistic Information in Speech using Two-Level Machine Learning Model

Velichko¹,

Saint-Petersburg²,

Karpov³

2021

Computational Linguistics and Intellectual Technologies

View full text Add to dashboard Cite

show abstract

“…Other research previously published concerning the RLDD database for DSD uses a speaker-level [ 23 ] or recording-level approach [ 24 , 26 , 27 ]; the former involves determining the overall truthful or deceptive attitude of each speaker (56 instances), while the latter attempts to classify each audio recording (121 total instances) in its entirety as truthful or deceptive . For the first case, the approach used in [ 23 ] reaches an accuracy of 61.0% using an MLP-based system and 52 audio features, and 71.2% using an RF of 100 trees with only the standard deviation of the pitch as input.…”

Section: Experimental Setup and Resultsmentioning

confidence: 99%

“…For the DSD task, previous research was carried out using traditional algorithmically extracted speech features or statistical descriptors derived from that, including the mean and standard deviation of the pitch [ 23 ], the MFCCs and their delta and double-delta coefficients [ 24 , 25 ], jitter, the harmonic-to-noise ratio (HNR) [ 26 ], or other acoustic and prosodic features [ 25 , 27 ] based on the ComParE feature set [ 28 ]. As for the machine learning models used, these include support vector machines (SVMs) [ 16 , 23 , 25 , 26 ], random forests (RFs) [ 20 , 23 , 25 , 27 , 29 , 30 ], MLPs [ 23 , 25 , 27 , 31 ], logistic regression [ 25 , 31 ], or ensemble methods with multiple classifiers and average/majority voting [ 27 ]. Specific results previously reported in literature for the databases used in this work for the DSD task, RLDD and RODeCAR, are provided in Section 3.5 .…”

Section: Introductionmentioning

confidence: 99%

Using Voice Activity Detection and Deep Neural Networks with Hybrid Speech Feature Extraction for Deceptive Speech Detection

Mihalache

Burileanu

2022

Sensors

View full text Add to dashboard Cite

In this work, we first propose a deep neural network (DNN) system for the automatic detection of speech in audio signals, otherwise known as voice activity detection (VAD). Several DNN types were investigated, including multilayer perceptrons (MLPs), recurrent neural networks (RNNs), and convolutional neural networks (CNNs), with the best performance being obtained for the latter. Additional postprocessing techniques, i.e., hysteretic thresholding, minimum duration filtering, and bilateral extension, were employed in order to boost performance. The systems were trained and tested using several data subsets of the CENSREC-1-C database, with different simulated ambient noise conditions, and additional testing was performed on a different CENSREC-1-C data subset containing actual ambient noise, as well as on a subset of the TIMIT database. An accuracy of up to 99.13% was obtained for the CENSREC-1-C datasets, and 97.60% for the TIMIT dataset. We proceed to show how the final VAD system can be adapted and employed within an utterance-level deceptive speech detection (DSD) processing pipeline. The best DSD performance is achieved by a novel hybrid CNN-MLP network leveraging a fusion of algorithmically and automatically extracted speech features, and reaches an unweighted accuracy (UA) of 63.7% on the RLDD database, and 62.4% on the RODeCAR database.

show abstract

Speech Deception Detection Based on EMD and Temporal Neural Network

Jiang

Chen

Yuan

et al. 2023

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Deceptive behaviour is a common phenomenon in human society. Research has shown that humans are not good at distinguishing deception, so studying automated deception detection techniques is a critical task. Most of the relevant technologies are susceptible to personal and environmental influences: EEG-based technologies need large and expensive equipment, facial-based technologies are sensitive with the camera’s perspective, and these reasons have somewhat limited the development of applications for deception detection technologies. In contrast, the equipment required for speech deception detection is cheap and easy to use, and the capture of speech is highly covert. Based on the application of signal decomposition algorithms in other fields such as EEG signals and speech emotion recognition, this paper proposed a signal decomposition and reconstruction method based on EMD to process the speech signal and a better deception detection performance was obtained by improving the speech quality. The comparison results with other decomposition algorithms showed that the EMD decomposition algorithm is the most suitable for our method. Across many different classification algorithms, accuracy improved by an average of 2.05% and the F1 score improved by an average of 1.7%. In addition, a new deception detector, called the TCN-LSTM network, was proposed in this paper. Experiments showed that this network organically combines the processing capability of TCN and LSTM for time series data; the recognition rate of deception detection was greatly improved, with the highest accuracy and F1 score reaching 86.2% and 86.0% under the EMD-based signal decomposition reconstruction method. Based on the research in this paper, the signal decomposition algorithms need to be further optimised for speech signals and more classification algorithms not used for this task should be tried.

show abstract

Applying Ensemble Learning Techniques and Neural Networks to Deceptive and Truthful Information Detection Task in the Flow of Speech

Cited by 3 publications

References 10 publications

Automatic Detection of Deceptive and Truthful Paralinguistic Information in Speech using Two-Level Machine Learning Model

Automatic Detection of Deceptive and Truthful Paralinguistic Information in Speech using Two-Level Machine Learning Model

Using Voice Activity Detection and Deep Neural Networks with Hybrid Speech Feature Extraction for Deceptive Speech Detection

Speech Deception Detection Based on EMD and Temporal Neural Network

Contact Info

Product

Resources

About