Detecting real life anger

Burkhardt, Felix; Polzehl, Tim; Stegmann, Joachim; Metze, Florian; Huber, Richard

doi:10.1109/icassp.2009.4960695

Cited by 44 publications

(28 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A Kappa value of 0 means no agreement, values between 0.4 and 0.7 are usually regarded as fair agreement and values above denote excellent agreement. Our labelers reached Kappa values between 0.79 for an early data collection [4] and 0.55 for the data set described here [6]. This indicates (not surprisingly) that agreement depends strongly on the data as well as the labelers.…”

Section: Data Labelingsupporting

confidence: 60%

See 1 more Smart Citation

Emotion detection in dialog systems: Applications, strategies and challenges

Burkhardt

Ballegooy

Engelbrecht

et al. 2009

2009 3rd International Conference on Affective Computing and Intelligent Interaction and Workshops

Self Cite

View full text Add to dashboard Cite

Emotion plays an important role in human communication and therefore also human machine dialog systems can benefit from affective processing. We present in this paper an overview of our work from the past few years and discuss general considerations, potential applications and experiments that we did with the emotional classification of human machine dialogs. Anger in voice portals as well as problematic dialog situations can be detected to some degree, but the noise in real life data and the issue of unambiguous emotion definition are still challenging. Also, a dialog system reacting emotionally might raise expectations with respect to its intellectual abilities that it can not fulfill.

show abstract

Section: Data Labelingsupporting

confidence: 60%

“…The progress in this work was reported in [7,4,17,5,6]. During this period, we experimented with different acoustic feature sets and different classifier algorithms.…”

Section: • Believable Agents Artificial Humansmentioning

confidence: 99%

Emotion detection in dialog systems: Applications, strategies and challenges

Burkhardt

Ballegooy

Engelbrecht

et al. 2009

2009 3rd International Conference on Affective Computing and Intelligent Interaction and Workshops

Self Cite

View full text Add to dashboard Cite

show abstract

“…Previous studies on both corpora yielded a much lower performance compared to our new findings. The former system described in (Schmitt et al, 2009) with the English database reached 72.6% f1 while the system described in (Burkhardt et al, 2009) developed for the German database reached 70% f1. The performance gain on the training set of respectively 5.6% and 4.7% f1 in our study can be attributed to the employment of the enhanced feature sets and the feature selection by IGR filtering.…”

Section: 3mentioning

confidence: 99%

“…For each turn, 3 labelers assigned one of the following labels: not angry, not sure, slightly angry, clear anger, clear rage or marked the turns as non applicable when encountering garbage. The labels were mapped onto two cover classes by clustering according to a threshold over the average of all voters' labels as described in (Burkhardt et al, 2009). Following Davies extension of Cohen's Kappa (Davies and Fleiss, 1982) for multiple labelers we obtain a value of = 0.52 which corresponds to moderate inter labeler agreement (Steidl et al, 2005).…”

Section: Selected Corporamentioning

confidence: 99%

Salient Features for Anger Recognition in German and English IVR Portals

Polzehl

Schmitt

Metze

2010

Spoken Dialogue Systems Technology and Design

Self Cite

View full text Add to dashboard Cite

Anger recognition in speech dialogue systems can help to enhance human computer interaction. In this paper we report on the setup and performance optimization techniques for successful anger classification using acoustic cues. We evaluate the performance of a broad variety of features on both a German and an American English voice portal database which contain "real" speech, i.e. non-acted, continuous speech of narrow-band quality. Starting with a large-scale feature extraction, we determine optimal sets of feature combinations for each language, by applying an Information-Gain based ranking scheme. Analyzing the ranking we notice that a large proportion of the most promising features for both databases are derived from MFCC and loudness. In contrast to this similarity also pitch features proved importance for the English database. We further calculate classification scores for our setups using discriminative training and Support-Vector Machine classification. The developed systems show that anger recognition in both English and Salient Features for Anger Recognition in German and English IVR PortalsGerman language can be processed very similarly reaching comparable results.

show abstract

“…Anger was detected in recordings from a German voice portal in [6]. Support vector machines (SVMs) and Gaussian mixture model (GMM)-based classifiers were applied to pitch, energy, duration, and spectral-related features.…”

Section: Introductionmentioning

confidence: 99%

Speaker-independent negative emotion recognition

Kotti

Paternò

Kotropoulos³

2010

2010 2nd International Workshop on Cognitive Information Processing

View full text Add to dashboard Cite

Abstract-This work aims to provide a method able to distinguish between negative and non-negative emotions in vocal interaction. A large pool of 1418 features is extracted for that purpose. Several of those features are tested in emotion recognition for the first time. Next, feature selection is applied separately to male and female utterances. In particular, a bidirectional Best First search with backtracking is applied. The first contribution is the demonstration that a significant number of features, first tested here, are retained after feature selection. The selected features are then fed as input to support vector machines with various kernel functions as well as to the K nearest neighbors classifier. The second contribution is in the speaker-independent experiments conducted in order to cope with the limited number of speakers present in the commonly used emotion speech corpora. Speakerindependent systems are known to be more robust and present a better generalization ability than the speaker-dependent ones. Experimental results are reported for the Berlin emotional speech database. The best performing classifier is found to be the support vector machine with the Gaussian radial basis function kernel. Correctly classified utterances are 86.73%±3.95% for male subjects and 91.73%±4.18% for female subjects. The last contribution is in the statistical analysis of the performance of the support vector machine classifier against the K nearest neighbors classifier as well as the statistical analysis of the various support vector machine kernels impact.

show abstract

Detecting real life anger

Cited by 44 publications

References 5 publications

Emotion detection in dialog systems: Applications, strategies and challenges

Emotion detection in dialog systems: Applications, strategies and challenges

Salient Features for Anger Recognition in German and English IVR Portals

Speaker-independent negative emotion recognition

Contact Info

Product

Resources

About