Natural human-computer interaction and audio-visual human behaviour sensing systems, which would achieve robust performance in-the-wild are more needed than ever as digital devices are becoming indispensable part of our life more and more. Accurately annotated real-world data are the crux in devising such systems. However, existing databases usually consider controlled settings, low demographic variability, and a single task. In this paper, we introduce the SEWA database of more than 2000 minutes of audio-visual data of 398 people coming from six cultures, 50% female, and uniformly spanning the age range of 18 to 65 years old. Subjects were recorded in two different contexts: while watching adverts and while discussing adverts in a video chat. The database includes rich annotations of the recordings in terms of facial landmarks, facial action units (FAU), various vocalisations, mirroring, and continuously valued valence, arousal, liking, agreement, and prototypic examples of (dis)liking. This database aims to be an extremely valuable resource for researchers in affective computing and automatic human sensing and is expected to push forward the research in human behaviour analysis, including cultural studies. Along with the database, we provide extensive baseline experiments for automatic FAU detection and automatic valence, arousal and (dis)liking intensity estimation.
Automatically classifying bird species by their sound signals is of crucial relevance for the research of ornithologists and ecologists. In this study, we present a novel framework for bird sounds classification from audio recordings. Firstly, the p-centre is used to detect the `syllables' of bird songs, which are the units for the recognition task; then, we use our openSMILE toolkit to extract large scales of acoustic features from chunked units of analysis (the `syllables'). ReliefF helps to reduce the dimension of the feature space. Lastly, an Extreme Learning Machine (ELM) serves for decision making. Results demonstrate that our system can achieve an excellent and robust performance scalable to different numbers of species (mean unweighted average recall of 93.82%, 89.56%, 85.30%, and 83.12% corresponding to 20, 30, 40, and 50 species of birds, respectively).