A5071848456- Speech Recognition and Synthesis
- Speech and Audio Processing
- Phonetics and Phonology Research
- Music and Audio Processing
- Syntax, Semantics, Linguistic Variation
- Emotion and Mood Recognition
- Natural Language Processing Techniques
- Face recognition and analysis
- Linguistic Variation and Morphology
- Speech and dialogue systems
- Advanced Data Compression Techniques
- Biometric Identification and Security
- Neurobiology of Language and Bilingualism
- Face and Expression Recognition
- Language, Discourse, Communication Strategies
- Topic Modeling
- Infant Health and Development
- EEG and Brain-Computer Interfaces
- Digital Media Forensic Detection
- Multisensory perception and integration
- Historical Economic and Social Studies
- Fuzzy Logic and Control Systems
- Innovation Policy and R&D
- Bayesian Methods and Mixture Models
- Language Development and Disorders
Medical University of Vienna
2024-2025
German Center for Neurodegenerative Diseases
2025
University Hospital Bonn
2025
Australian National University
1996-2024
University of Canberra
2010-2023
McGill University
2013-2023
Western University
2015
Cornell University
2006-2015
Deutsche Nationalbibliothek
2015
University of Kassel
2014
This paper reports three studies aimed at addressing questions about the acoustic correlates of information structure in English: (1) do speakers mark prosodically, and, to extent they do; (2) what are features associated with different aspects structure; and (3) how well can listeners retrieve this from signal? The subject–verb–object sentences was manipulated via preceding those sentences: elements target were either focused (i.e., answer a wh-question) or given mentioned prior discourse);...
Emotion recognition is a very active field of research. The Recognition In Wild Challenge and Workshop (EmotiW) 2013 Grand consists an audio-video based emotion classification challenges, which mimics real-world conditions. Traditionally, has been performed on laboratory controlled data. While undoubtedly worthwhile at the time, such data poorly represents environment conditions faced in situations. goal this to define common platform for evaluation methods database challenge Acted Facial...
The analysis of contrastive topics introduced in Büring 1997b and further developed 2003 relies on distinguishing two types constituents that introduce alternatives: the sentence focus, which is marked by a FOC feature, topic, CT feature. A non-compositional rule interpretation refers to these features used derive topic semantic value, nested set sets propositions. This paper presents evidence for correlation between restrictive syntax focus operators topics, unexpected under this analysis....
An estimated 350 million people worldwide are affected by depression. Using affective sensing technology, our long-term goal is to develop an objective multimodal system that augments clinical opinion during the diagnosis and monitoring of This paper steps towards developing a classification system-oriented approach, where feature selection, fusion-based experiments conducted infer which types behaviour (verbal nonverbal) combinations can best discriminate between depression non-depression....
Depression is a common and disabling mental health disorder, which impacts not only on the sufferer but also their families, friends economy overall. Despite its high prevalence, current diagnosis relies almost exclusively patient self-report clinical opinion, leading to number of subjective biases. Our aim develop an objective affective sensing system that supports clinicians in monitoring depression. In this paper, we analyse performance eye movement features extracted from face videos...
Depression is a common and disabling mental health disorder, which impacts not only on the sufferer but also their families, friends economy overall. Our ultimate aim to develop an automatic objective affective sensing system that supports clinicians in diagnosis monitoring of clinical depression. Here, we analyse performance head pose movement features extracted from face videos using 3D model projected 2D Active Appearance Model (AAM). In binary classification task (depressed vs....
Major depressive disorders are mental of high prevalence, leading to a impact on individuals, their families, society and the economy. In order assist clinicians better diagnose depression, we investigate an objective diagnostic aid using affective sensing technology with focus acoustic features. this paper, hypothesise that (1) classifying general characteristics clinical depression spontaneous speech will give results than read speech, (2) there some features robust would good...
No abstract.
We consider mimicry, a simple technology form of attack requiring low level expertise, to investigate whether speaker recognition system is vulnerable mimicry by an impostor without using the assistance any other technologies. Experiments on 138 speakers in YOHO database and two people who played role as imitators have shown that can if knows registered has very similar voice impostor's voice.
Accurate detection of depression from spontaneous speech could lead to an objective diagnostic aid assist clinicians better diagnose depression. Little thought has been given so far which classifier performs best for this task. In study, using a 60-subject real-world clinically validated dataset, we compare three popular classifiers the affective computing literature - Gaussian Mixture Models (GMM), Support Vector Machines (SVM) and Multilayer Perceptron neural networks (MLP) as well...
Many phonological processes can be affected by segmental context spanning word boundaries, which often lead to variable outcomes. This paper tests the idea that some of this variability explained reference production planning. We examine coronal stop deletion (CSD), a process conditioned preceding and upcoming context, in corpus spontaneous British English speech, as means investigating number variables associated with planning: Prosodic boundary strength, frequency, conditional probability...
Millions of people worldwide suffer from depression. Do commonalities exist in their nonverbal behavior that would enable cross-culturally viable screening and assessment severity? We investigated the generalisability an approach to detect depression severity using video-recorded clinical interviews Australia, USA Germany. The material varied type interview, subtypes inclusion healthy control subjects, cultural background, recording environment. analysis focussed on temporal features...