Vocal emotion recognition (VER) in natural speech, often referred to as speech emotion recognition (SER), remains challenging for both humans and computers. Applied fields including clinical diagnosis and intervention, social interaction research or Human Computer Interaction (HCI) increasingly benefit from efficient VER algorithms. Several feature sets were used with machine-learning (ML) algorithms for discrete emotion classification. However, there is no consensus for which low-level-descriptors and classifiers are optimal. Therefore, we aimed to compare the performance of machine-learning algorithms with several different feature sets. Concretely, seven ML algorithms were compared on the Berlin Database of Emotional Speech: Multilayer Perceptron Neural Network (MLP), J48 Decision Tree (DT), Support Vector Machine with Sequential Minimal Optimization (SMO), Random Forest (RF), k-Nearest Neighbor (KNN), Simple Logistic Regression (LOG) and Multinomial Logistic Regression (MLR) with 10-fold cross validation using four openSMILE feature sets (i.e., IS-09, emobase, GeMAPS and eGeMAPS). Results indicated that SMO, MLP and LOG show better performance (reaching to 87.85%, 84.00% and 83.74% accuracies, respectively) compared to RF, DT, MLR and KNN (with minimum 73.46%, 53.08%, 70.65% and 58.69% accuracies, respectively). Overall, the emobase feature set performed best. We discuss the implications of these findings for applications in diagnosis, intervention or HCI.

Since COVID-19 has become a pandemic, everyday life has seen dramatic changes affecting individuals, families, and children with and without autism. Among other things, these changes entail more time at home, digital forms of communication, school closures, and reduced support and intervention. Here, we assess the effects of the pandemic on quality of life for school-age autistic and neurotypical children and adolescents. First, we provide a comprehensive review of the current relevant literature. Next, we report original data from a survey conducted in several countries, assessing activities, well-being, and social life in families with autism, and their changes over time. We focus on differences between children with and without autism from within the same families, and on different outcomes for children with high- or low-functioning autism. While individuals with autism scored lower in emotional and social functioning than their neurotypical siblings, both groups of children showed comparable decreases in well-being and increases in anxiety, compared to before the pandemic. By contrast, decreases in adaptability were significantly more pronounced in autistic children and adolescents compared to neurotypical children and adolescents. Overall, although individual families reported some positive effects of pandemic restrictions, our data provide no evidence that these generalize across children and adolescents with autism, or even just to individuals with high-functioning autism. We discuss the increased challenges that need to be addressed to protect children and adolescents’ well-being under pandemic conditions, but also point out potentials in the present situation that could be used towards social participation and success in older children and young adults with autism.