Combining Electrodermal Activity and Speech Analysis towards a more Accurate Emotion Recognition System

Greco, Alberto; Marzi, Claudia; Lanatà, Antonio; Scilingo, Enzo Pasquale; Vanello, Nicola

doi:10.1109/embc.2019.8857745

Cited by 14 publications

(9 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study, we analyzed only the stress stages during which the subjects were not talking to avoid any interaction between EDA signals and speech [66]. Indeed, speech induces physiological irregular respiration that activates the sympathetic reflex and consequently affects the sweat gland dynamics and the related EDA signal [66], [68]. Accordingly, we selected three stress stages: ST, SA, VE in addition to the basal stage (BS).…”

Section: Feature Extractionmentioning

confidence: 99%

Acute Stress State Classification Based on Electrodermal Activity Modeling

Greco

Valenza

Lázaro

et al. 2023

IEEE Trans. Affective Comput.

Self Cite

View full text Add to dashboard Cite

Acute stress is a physiological condition that may induce several neural dysfunctions with a significant impact on life quality. Accordingly, it would be important to monitor stress in everyday life unobtrusively and inexpensively. In this paper, we presented a new methodological pipeline to recognize acute stress conditions using electrodermal activity (EDA) exclusively. Particularly, we combined a rigorous and robust model (cvxEDA) for EDA processing and decomposition, with an algorithm based on a support vector machine to classify the stress state at a singlesubject level. Indeed, our method, based on a single sensor, is robust to noise, applies a rigorous phasic decomposition, and implements an unbiased multiclass classification. To this end, we analyzed the EDA of 65 volunteers subjected to different acute stress stimuli induced by a modified version of the Trier Social Stress Test. Our results show that stress is successfully detected with an average accuracy of 94.62%. Besides, we proposed a further 4-class pattern recognition system able to distinguish between non-stress condition and three different stressful stimuli achieving an average accuracy as high as 75.00%. These results, obtained under controlled conditions, are the first step towards applications in ecological scenarios.

show abstract

Section: Feature Extractionmentioning

confidence: 99%

Acute Stress State Classification Based on Electrodermal Activity Modeling

Greco

Valenza

Lázaro

et al. 2023

IEEE Trans. Affective Comput.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Physiological signals provide more continuous real-time monitoring compared to facial expressions. In comparable studies [28][29][30][31][32][33][34][35], the impact of using physiological signals for emotion detection and subsequent recognition is highlighted. Shukla J. et al (2021) [28] assessed and evaluated different techniques for EDA signals and determined the optimal number of features required to yield high accuracy and real-time emotion recognition.…”

Section: Related Workmentioning

confidence: 99%

“…A combination, more commonly known as fusion, of more than one signal for emotion recognition has also been studied, with promising results. Greco A. et al (2019) explored the fusion of both EDA signals and speech patterns to improve arousal level recognition, yielding a marginal classifier improvement of 11.64% using an SVM classifier with recursive feature elimination [32]. Du G. et al (2020) investigated the combination of facial expressions and HR for emotion recognition in gaming environments, increasing the recognition accuracy by 8.30% [33].…”

Section: Related Workmentioning

confidence: 99%

Harnessing Wearable Devices for Emotional Intelligence: Therapeutic Applications in Digital Health

Arabian,

Abdulbaki Alshirbaji,

Schmid

et al. 2023

Sensors

View full text Add to dashboard Cite

Emotional intelligence strives to bridge the gap between human and machine interactions. The application of such systems varies and is becoming more prominent as healthcare services seek to provide more efficient care by utilizing smart digital health apps. One application in digital health is the incorporation of emotion recognition systems as a tool for therapeutic interventions. To this end, a system is designed to collect and analyze physiological signal data, such as electrodermal activity (EDA) and electrocardiogram (ECG), from smart wearable devices. The data are collected from different subjects of varying ages taking part in a study on emotion induction methods. The obtained signals are processed to identify stimulus trigger instances and classify the different reaction stages, as well as arousal strength, using signal processing and machine learning techniques. The reaction stages are identified using a support vector machine algorithm, while the arousal strength is classified using the ResNet50 network architecture. The findings indicate that the EDA signal effectively identifies the emotional trigger, registering a root mean squared error (RMSE) of 0.9871. The features collected from the ECG signal show efficient emotion detection with 94.19% accuracy. However, arousal strength classification is only able to reach 60.37% accuracy on the given dataset. The proposed system effectively detects emotional reactions and can categorize their arousal strength in response to specific stimuli. Such a system could be integrated into therapeutic settings to monitor patients’ emotional responses during therapy sessions. This real-time feedback can guide therapists in adjusting their strategies or interventions.

show abstract

“…On average, RIPPER achieved 92.01% accuracy in the binary assessment of sad versus relaxed emotional state. Greco, et al (2019) investigated the possibility of combining electrodermal activity (EDA) and voice data to recognize human arousal levels during single effective word pronunciation. The support vector machine with a recursive feature elimination (SVM-RFE) was trained and tested on three datasets, using the two channels (speech and EDA) independently and combined.…”

Section: Related Workmentioning

confidence: 99%

Driver Drowsiness Detection Using Gray Wolf Optimizer Based on Voice Recognition

Jasim

Hassan

Turner

2022

ARO

View full text Add to dashboard Cite

Globally, drowsiness detection prevents accidents. Blood biochemicals, brain impulses, etc., can measure tiredness. However, due to user discomfort, these approaches are challenging to implement. This article describes a voice-based drowsiness detection system and shows how to detect driver fatigue before it hampers driving. A neural network and Gray Wolf Optimizer are used to classify sleepiness automatically. The recommended approach is evaluated in alert and sleep-deprived states on the driver tiredness detection voice real dataset. The approach used in speech recognition is mel-frequency cepstral coefficients (MFCCs) and linear prediction coefficients (LPCs). The SVM algorithm has the lowest accuracy (71.8%) compared to the typical neural network. GWOANN employs 13-9-7-5 and 30-20-13-7 neurons in hidden layers, where the GWOANN technique had 86.96% and 90.05% accuracy, respectively, whereas the ANN model achieved 82.50% and 85.27% accuracy, respectively.

show abstract

Combining Electrodermal Activity and Speech Analysis towards a more Accurate Emotion Recognition System

Cited by 14 publications

References 29 publications

Acute Stress State Classification Based on Electrodermal Activity Modeling

Acute Stress State Classification Based on Electrodermal Activity Modeling

Harnessing Wearable Devices for Emotional Intelligence: Therapeutic Applications in Digital Health

Driver Drowsiness Detection Using Gray Wolf Optimizer Based on Voice Recognition

Contact Info

Product

Resources

About