Enhancing Emotion Recognition from Speech through Feature Selection

Κωστούλας, Θεόδωρος; Ganchev, Todor; Lazaridis, Alexandros; Fakotakis, Nikos

doi:10.1007/978-3-642-15760-8_43

Cited by 21 publications

(18 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The bio-signal, facial (facial gestures) and the speech based emotion recognition (Kostoulas et al, 2010a, 2010b) provides detection of the player's emotions while confronted with specific game situations and the triggered emotions. Physiological reactivity and emotional recognition continuously track the emotional state of the player along the video game, while the game automatically responds in return by modifying aspects of the game play difficulty, in a closed loop.…”

Section: Introductionmentioning

confidence: 99%

Video games as a complementary therapy tool in mental disorders: PlayMancer, a European multicentre study

Fernández‐Aranda¹,

Jiménez‐Murcia²,

Santamaría³

et al. 2012

Journal of Mental Health

Self Cite

165

116

View full text Add to dashboard Cite

Background: Previous review studies have suggested that computer games can serve as an alternative or additional form of treatment in several areas (schizophrenia, asthma or motor rehabilitation). Although several naturalistic studies have been conducted showing the usefulness of serious video games in the treatment of some abnormal behaviours, there is a lack of serious games specially designed for treating mental disorders.Aim: The purpose of our project was to develop and evaluate a serious video game designed to remediate attitudinal, behavioural and emotional processes of patients with impulse-related disorders.Method and results: The video game was created and developed within the European research project PlayMancer. It aims to prove potential capacity to change underlying attitudinal, behavioural and emotional processes of patients with impulse-related disorders. New interaction modes were provided by newly developed components, such as emotion recognition from speech, face and physiological reactions, while specific impulsive reactions were elicited. The video game uses biofeedback for helping patients to learn relaxation skills, acquire better self-control strategies and develop new emotional regulation strategies. In this article, we present a description of the video game used, rationale, user requirements, usability and preliminary data, in several mental disorders.

show abstract

Section: Introductionmentioning

confidence: 99%

Video games as a complementary therapy tool in mental disorders: PlayMancer, a European multicentre study

Fernández‐Aranda¹,

Jiménez‐Murcia²,

Santamaría³

et al. 2012

Journal of Mental Health

Self Cite

165

116

View full text Add to dashboard Cite

show abstract

“…Energy Mean and MFCCs (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12) standard LPCs (0-13) deviation ZCR Spectral Flux Spectral Rolloff Chroma Vector (0-11) Clarity Table 1. Extracted audio features…”

Section: Low Level Descriptor Functionalsmentioning

confidence: 99%

“…Commonly extracted are the features related to pitch, formants, loudness, harmonic-to-noice-ratio, harmonic rations, jitter and shimmer [9]. Several classifiers have been used in the past, such as non-negative matrix factorisation [3], Gaussian Mixture Models [10], Hidden Markov Models [11], Support Vector Machines [12], Artificial Neural Networks (ANNs) either swallow or deep [13], Decision Trees or k-Nearest Neighbor distance classifiers [14].…”

Section: Introductionmentioning

confidence: 99%

Effective emotion recognition in movie audio tracks

Kotti

Stylianou

2017

2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

This paper addresses the problem of speech emotion recognition from movie audio tracks. The recently collected Acted Facial Expression in the Wild 5.0 database is used. The aim is to discriminate among angry, happy, and neutral. We extract a relatively small number of features, a subset of which is not commonly used for the emotion recognition task. Those features are fed as input to an ensemble classifier that combines random forests with support vector machines. An accuracy of 65.63% is reported, outperforming a baseline system that uses the K-nearest neighbor classifier and has an accuracy of 56.88%. To verify the suitability of the exploited features, the same ensemble classification schema is applied on the feature set similar those employed in Audio/Visual Emotion Challenge 2011. In the latter case, an accuracy of 61.25% is achieved using a large set of 1582 features, as opposed to just 86 features in our case that lead to a relative improvement of 7.15% in accuracy.

show abstract

“…One popular technique for selecting those features is Principal Component Analysis (PCA) which is extensively used in [6][7][8]. Correlation-based Sub Set Evaluators is the other famous method which improved many recognition systems such as in [9,10]. Other feature selection algorithms such as Mutual Information (MI) [7,11], Canonical Correlation Analysis (CCA) [12] and Sequential Floating Forward Selection (SFFS) algorithm [13].…”

Section: Related Workmentioning

confidence: 99%

Speech Emotion Recognition based on SVM as Both Feature Selector and Classifier

Shirani¹,

Nilchi²

2016

IJIGSP

View full text Add to dashboard Cite

Abstract-The aim of this paper is to utilize Support Vector Machine (SVM) as feature selection and classification techniques for audio signals to identify human emotional states. One of the major bottlenecks of common speech emotion recognition techniques is to use a huge number of features per utterance which could significantly slow down the learning process, and it might cause the problem known as -the curse of dimensionality‖. Consequently, to ease this challenge this paper aims to achieve high accuracy system with a minimum set of features. The proposed model uses two methods, namely -SVM features selection‖ and the common -Correlation-based Feature Subset Selection (CFS)‖ for the feature dimensions reduction part. In addition, two different classifiers, one Support Vector Machine and the other Neural Network are separately adopted to identify the six emotional states of anger, disgust, fear, happiness, sadness and neutral. The method has been verified using Persian (Persian ESD) and German (EMO-DB) emotional speech databases, which yield high recognition rates in both databases. The results show that SVM feature selection method provides better emotional speech-recognition performance compared to CFS and baseline feature set. Moreover, the new system is able to achieve a recognition rate of (99.44%) on the Persian ESD and (87.21%) on Berlin Emotion Database for speaker-dependent classification. Besides, promising result (76.12%) is obtained for speaker-independent classification case; which is among the best-known accuracies reported on the mentioned database relative to its little number of features.

show abstract

Enhancing Emotion Recognition from Speech through Feature Selection

Cited by 21 publications

References 15 publications

Video games as a complementary therapy tool in mental disorders: PlayMancer, a European multicentre study

Video games as a complementary therapy tool in mental disorders: PlayMancer, a European multicentre study

Effective emotion recognition in movie audio tracks

Speech Emotion Recognition based on SVM as Both Feature Selector and Classifier

Contact Info

Product

Resources

About