@inproceedings{SchullerVlasenkoEybenetal.2015, author = {Schuller, Bj{\"o}rn and Vlasenko, Bogdan and Eyben, Florian and W{\"o}llmer, Martin and Stuhlsatz, Andr{\´e} and Wendemuth, Andreas and Rigoll, Gerhard}, title = {Cross-Corpus Acoustic Emotion Recognition: Variances and Strategies (Extended Abstract)}, series = {In Proc. 6th biannual Conference on Affective Computing and Intelligent Interaction (ACII 2015), AAAC,}, booktitle = {In Proc. 6th biannual Conference on Affective Computing and Intelligent Interaction (ACII 2015), AAAC,}, year = {2015}, abstract = {As the recognition of emotion from speech has matured to a degree where it becomes applicable in real-life settings, it is time for a realistic view on obtainable performances. Most studies tend to overestimation in this respect: acted data is often used rather than spontaneous data, results are reported on pre-selected prototypical data, and true speaker disjunctive partitioning is still less common than simple cross-validation. A considerably more realistic impression can be gathered by inter-set evaluation: we therefore show results employing six standard databases in a cross-corpora evaluation experiment. To better cope with the observed high variances, different types of normalization are investigated. 1.8k individual evaluations in total indicate the crucial performance inferiority of inter- to intra-corpus testing.}, language = {en} } @inproceedings{StuhlsatzMeyerEybenetal.2011, author = {Stuhlsatz, Andr{\´e} and Meyer, Christine and Eyben, Florian and Zielke, Thomas and Meier, Gunter and Schuller, Bjorn}, title = {Deep neural networks for acoustic emotion recognition: Raising the benchmarks}, series = {2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, booktitle = {2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, publisher = {IEEE}, doi = {10.1109/ICASSP.2011.5947651}, pages = {5688 -- 5691}, year = {2011}, abstract = {Deep Neural Networks (DNNs) denote multilayer artificial neural networks with more than one hidden layer and millions of free parameters. We propose a Generalized Discriminant Analysis (GerDA) based on DNNs to learn discriminative features of low dimension optimized with respect to a fast classification from a large set of acoustic features for emotion recognition. On nine frequently used emotional speech corpora, we compare the performance of GerDA features and their subsequent linear classification with previously reported benchmarks obtained using the same set of acoustic features classified by Support Vector Machines (SVMs). Our results impressively show that low-dimensional GerDA features capture hidden information from the acoustic features leading to a significantly raised unweighted average recall and considerably raised weighted average recall.}, language = {en} } @article{SchullerVlasenkoEybenetal.2010, author = {Schuller, Bjorn and Vlasenko, Bogdan and Eyben, Florian and Wollmer, Martin and Stuhlsatz, Andr{\´e} and Wendemuth, Andreas and Rigoll, Gerhard}, title = {Cross-Corpus Acoustic Emotion Recognition: Variances and Strategies}, series = {IEEE Transactions on Affective Computing}, volume = {1}, journal = {IEEE Transactions on Affective Computing}, number = {2}, publisher = {IEEE}, doi = {10.1109/t-affc.2010.8}, pages = {119 -- 131}, year = {2010}, language = {en} }