Alicia Flores Requardt

Bio: Alicia Flores Requardt is an academic researcher from Otto-von-Guericke University Magdeburg. The author has contributed to research in topics: Multimodal interaction & Room acoustics. The author has an hindex of 1, co-authored 4 publications receiving 7 citations.

Towards affect-aware vehicles for increasing safety and comfort: recognising driver emotions from audio recordings in a realistic driving study

Abstract: For vehicle safety, the in-time monitoring of the driver and assessing his/her state is a demanding issue. Frustration can lead to aggressive driving behaviours, which play a decisive role in up to one-third of fatal road accidents. Consequently, the authors present the automatic analysis of the emotional driver states of frustration, anxiety, positive and neutral. Based on experiments with normal drivers within cars in real-world (low expressivity) situations, they use speech data, as speech can be recorded with zero invasiveness and comes naturally in driving situations. A careful selection of speech features, subject data identification, hyper-parameter optimisation, and machine learning algorithms was applied for this difficult 4-emotion-class detection problem, where the literature hardly reports results above chance level. In-car assistance demands real-time computing. A very detailed analysis yields best results with relatively small random forests, and with an optimal feature set containing only 65 features (6.51% of the standard emobase feature set) which outperformed all other feature sets, producing 35.38% unweighted average recall (53.26% precision) with low computational effort, and also reducing the inevitably high confusion of ‘neutral’ with low-expressed emotions. This result is comparable to and even outperforming other reported studies of emotion recognition in the wild. Their work, therefore, triggers adaptive automotive safety applications.

Anticipating the User: Acoustic Disposition Recognition in Intelligent Interactions

Abstract: Contemporary technical devices obey the paradigm of naturalistic multimodal interaction and user-centric individualisation. Users expect devices to interact intelligently, to anticipate their needs, and to adapt to their behaviour. To do so, companion-like solutions have to take into account the affective and dispositional state of the user, and therefore to be trained and modified using interaction data and corpora. We argue that, in this context, big data alone is not purposeful, since important effects are obscured, and since high-quality annotation is too costly. We encourage the collection and use of enriched data. We report on recent trends in this field, presenting methodologies for collecting data with rich disposition variety and predictable classifications based on a careful design and standardised psychological assessments. Besides socio-demographic information and personality traits, we also use speech events to improve user state models. Furthermore, we present possibilities to increase the amount of enriched data in cross-corpus or intra-corpus way based on recent learning approaches. Finally, we highlight particular recent neural recognition approaches feasible for smaller datasets, and covering temporal aspects.

Analysis of the influence of different room acoustics on acoustic emotion features

Abstract: In automatic analyses of speech and emotion recognition, it has to beensured that training and test conditions are similar. The presented study aims toinvestigate the influence of certain room acoustics on common features used foremotion recognition. As a benchmark database this study focuses on the BerlinDatabase of Emotional Speech. The following rooms were analysed: a) modernlecture hall, b) older lecture hall, and c) staircase. For all rooms and their differentrecording setups, different acoustic measures were captured. The speech record-ings analysed in this paper were realized only at the ideal locations within therooms. Afterwards, 52 features (LLDs of emobase) were automatically extractedusing OpenSMILE and a sample-wise statistical analysis (pairedt-test) was carriedout. Therefore, the number of acoustically degraded features and its effect sizecan be linked to the acoustic parameters of the different recording experiments. Asresult, 15% of the degraded samples show a highly significant difference regard-ing all considered rooms. Especially MFCCs account for approximate 50% of thedegradation. Furthermore, the degradation is analysed depending on the emotionand room acoustic.

An Experimental Paradigm for Inducing Emotions in a Real World Driving Scenario Evidence from Self-Report, Annotation of Speech Data and Peripheral Physiology

Abstract: Empathic vehicles are a promising concept to increase the safety and acceptance of automated vehicles. However, on the way towards empathic vehicles a lot of research in the area of automated emotion recognition is necessary. Successful methods to detect emotions need to be trained on realistic data that contain the target emotion and come from a setting close to the final application. At the moment, data sets fulfilling these requirements are lacking. Therefore, the goal of this work is to present an experimental paradigm that induces four different emotional states (neutral, positive, frustration and mild anxiety) in a real-world driving setting using a combination of secondary tasks and conversation-based emotional recall. An evaluation of the paradigm using self-report data, annotation of speech data and peripheral physiology indicates that the methods to induce the target emotions were successful. Based on the insights of the experiment, finally a list of recommendations for the induction of emotions in real world driving settings is given.

In-Vehicle Sensing for Smart Cars

Abstract: Driving safety has been attracting more and more interest due to the unprecedented proliferation of vehicles and the subsequent increase of traffic accidents. As such the research community has been actively seeking solutions that can make vehicles more intelligent and thus improve driving safety in everyday life. Among all the existing approaches, in-vehicle sensing has become a great preference by monitoring the driver’s health, emotion, attention, etc., which can offer rich information to the advanced driving assistant systems (ADAS) to respond accordingly and thus reduce injuries as much/early as possible. There have been many significant developments in the past few years on in-vehicle sensing. The goal of this paper is to provide a comprehensive review of the motivation, applications, state-of-the-art developments, and possible future interests in this research area. According to the application scenarios, we group the existing works into five categories, including occupancy detection, fatigue/drowsiness detection, distraction detection, driver authentication, and vital sign monitoring, review the fundamental techniques adopted, and present their limitations for further improvement. Finally, we discuss several future trends for enhancing current capabilities and enabling new opportunities for in-vehicle sensing.

A Comprehensive Review of Driving Style Evaluation Approaches and Product Designs Applied to Vehicle Usage-Based Insurance

Abstract: Vehicle insurance is a very important source of income for insurance companies, and it is closely related to the driving style performed by driving behavior. Different driving styles can better reflect the driving risk than the number of violations, claims, and other static statistic data. Subdivide the vehicle insurance market according to the personal characteristics and driving habits of the insured vehicles, and studying the personalized vehicle insurance products, will help the insurance companies to improve their income, help the drivers to change their bad driving habits, and thus help to realize the healthy development of the vehicle insurance industry. In the past 20 to 30 years, more and more insurance companies around the world have launched vehicle usage-based insurance (UBI) products based on driving style analysis. However, up to now, there are few comprehensive reports on commercial vehicle UBI products and their core driving risk assessment methods. On the basis of literature indexing on the Web of Science and other academic platforms by using the keywords involved in vehicle UBI, over 100 relevant works of literature were screened in this paper, and a detailed and comprehensive discussion on the driving style evaluation methods and the design of commercial vehicle UBI products during the past 20 to 30 years has been made, hoping to get a full understanding of the possible factors affecting driving style and the collectible data that can reflect these factors, and to get a full grasp of the developing status, challenges and future trends in vehicle insurance branch of the Internet of Vehicles (IoV) industry.

Modelling Contexts for Interactions in Dynamic Open-World Scenarios

Abstract: Intelligent Interactive Systems work well in well defined contexts. Therefore, current research focuses on developing specialised systems for specific tasks. However, real life situations violate these restrictions. Even with a specialised task focus Intelligent Interactive Systems still need to be able to react in a meaningful way in real world situations. In this paper, we propose a set of context factors to determine context and a coarse dependency structure for predicting behaviour. We argue that with such a limited set of context factors it will be possible to model context-specific behaviours for different contexts. Also, it can serve as a basis to determine the complexity of a situation.