Emotion Recognition in Human-Computer Interaction

The authors are concerned with the development of computer systems that are capable of using information from faces and voices to recognise people's emotions in real-life situations. The paper addresses the nature of the challenges that lie ahead, and provides an assessment of the progress that has been made in the areas of signal processing and analysis techniques (with regard to speech and face), and the psychological and linguistic analyses of emotion. Ongoing developmental work by the authors in each of these areas is described.

On-line Emotion Recognition in a 3-D Activation-Valence-Time Continuum using Acoustic and Linguistic Cues

For many applications of emotion recognition, such as virtual agents, the system must select responses while the user is speaking. This requires reliable on-line recognition of the user’s affect. However most emotion recognition systems are based on turnwise processing. We present a novel approach to on-line emotion recognition from speech using Long Short-Term Memory Recurrent Neural Networks. Emotion is recognised frame-wise in a two-dimensional valence-activation continuum. In contrast to current state-of-the-art approaches, recognition is performed on low-level signal frames, similar to those used for speech recognition. No statistical functionals are applied to low-level feature contours. Framing at a higher level is therefore unnecessary and regression outputs can be produced in real-time for every low-level input frame. We also investigate the benefits of including linguistic features on the signal frame level obtained by a keyword spotter.

AVEC 2011–The First International Audio/Visual Emotion Challenge

The Audio/Visual Emotion Challenge and Workshop (AVEC 2011) is the first competition event aimed at comparison of multimedia processing and machine learning methods for automatic audio, visual and audiovisual emotion analysis, with all participants competing under strictly the same conditions. This paper first describes the challenge participation conditions. Next follows the data used – the SEMAINE corpus – and its partitioning into train, development, and test partitions for the challenge with labelling in four dimensions, namely activity, expectation, power, and valence. Further, audio and video baseline features are introduced as well as baseline results that use these features for the three sub-challenges of audio, video, and audiovisual emotion recognition.

The relationship between the recognition of facial expressions and self-reported anger in people with intellectual disabilities

Background This study aims to examine the relationship between how individuals with intellectual disabilities report their own levels of anger, and the ability of those individuals to recognize emotions. It was hypothesized that increased expression of anger would be linked to lower ability to recognize facial emotional expressions and increased tendency to interpret facial expressions in a hostile or negative manner. It was also hypothesized increased levels of anger may lead to the altered perception of a particular emotion.

Method A cross-sectional survey design was used. Thirty participants completed a test of facial emotion recognition (FER), and a self-report anger inventory (Benson & Ivins 1992) as part of a structured interview.

Results Individuals with higher self-reported anger did not show significantly reduced performance in FER, or interpret facial expressions in a more hostile manner compared with individuals with less self-reported anger. However, they were less accurate in recognizing neutral facial emotions.

Conclusions It is tentatively suggested that individuals with high levels of anger may be likely to perceive emotional content in a neutral facial expression because of their high levels of emotional arousal.

Laughter Type Recognition from Whole Body Motion

Despite the importance of laughter in social interactions it remains little studied in affective computing. Respiratory, auditory, and facial laughter signals have been investigated but laughter-related body movements have received almost no attention. The aim of this study is twofold: first an investigation into observers' perception of laughter states (hilarious, social, awkward, fake, and non-laughter) based on body movements alone, through their categorization of avatars animated with natural and acted motion capture data. Significant differences in torso and limb movements were found between animations perceived as containing laughter and those perceived as nonlaughter. Hilarious laughter also differed from social laughter in the amount of bending of the spine, the amount of shoulder rotation and the amount of hand movement. The body movement features indicative of laughter differed between sitting and standing avatar postures. Based on the positive findings in this perceptual study, the second aim is to investigate the possibility of automatically predicting the distributions of observer's ratings for the laughter states. The findings show that the automated laughter recognition rates approach human rating levels, with the Random Forest method yielding the best performance.

The Belfast Induced Natural Emotion Database

For many years psychological research on facial expression of emotion has relied heavily on a recognition paradigm based on posed static photographs. There is growing evidence that there may be fundamental differences between the expressions depicted in such stimuli and the emotional expressions present in everyday life. Affective computing, with its pragmatic emphasis on realism, needs examples of natural emotion. This paper describes a unique database containing recordings of mild to moderate emotionally coloured responses to a series of laboratory based emotion induction tasks. The recordings are accompanied by information on self-report of emotion and intensity, continuous trace-style ratings of valence and intensity, the sex of the participant, the sex of the experimenter, the active or passive nature of the induction task and it gives researchers the opportunity to compare expressions from people from more than one culture.

Empowering the Shamed Self: Recognition and Critical Social Work

Summary: This article provides a review of the contribution of Axel Honneth’s model of recognition for critical social work. While Honneth’s tripartite conceptualisation of optimal identity-formation is positively appraised, his analysis of the link between misrecognition, the experience of shame and eventual sense of moral outrage, is contested. Drawing on a range of sources, including the sociology of shame, Honneth’s ideas about the emotional antecedents of emancipatory action are revised to guide critical social work with misrecognised service users.

Findings: The intellectual background to Honneth’s recognition model, emanating from leading German philosophers, is described and its application to social work set out. Even so, Honneth’s model is found to be deficient in one primary regard: its assumption about the emotional antecedents to quests for withheld recognition is misapprehended. In particular, the argument in this article is that the ubiquitous emotion of shame, which Honneth argues flows from misrecognition, must be carefully addressed through the medium of relationship, otherwise it might lead to repressed shame and frustrated attempts at social struggle. To this end, a social work process is delineated for dealing with shame, following episodes of misrecognition.

Applications: Honneth’s model of recognition, along with revised ideas about how to recognise and manage shame, is incorporated into a conceptual framework for critical social work practice. With this renewed understanding of the impact of shame, following misrecognition, social workers should be better equipped conceptually to enable service users to take action for empowerment.