A5002505718- Speech Recognition and Synthesis
- Sentiment Analysis and Opinion Mining
- Music and Audio Processing
- Emotion and Mood Recognition
- Voice and Speech Disorders
- Speech and Audio Processing
- Topic Modeling
- Speech and dialogue systems
- Text and Document Classification Technologies
- Video Analysis and Summarization
- Enzyme Catalysis and Immobilization
- Microbial Metabolic Engineering and Bioproduction
- Image Retrieval and Classification Techniques
- Mental Health via Writing
- Advanced Text Analysis Techniques
- Advanced Image and Video Retrieval Techniques
- Biofuel production and bioconversion
- Phonetics and Phonology Research
University of Lisbon
2018-2022
Institute for Biotechnology and Bioengineering
2022
Carnegie Mellon University
2016-2021
Instituto de Engenharia de Sistemas e Computadores Investigação e Desenvolvimento
2018-2021
Instituto Superior Técnico
2019
The discovery of itaconic acid as a product citric pyrolytic distillation in 1837 opened the possibility using it polymer building block. Itaconic acid, featuring two carboxylic acids and an unsaturated group, can potentially be used block several chemical syntheses, with particular emphasis on manufacture. elucidation biochemical pathways originating from first Aspergillus terreus and, recently, species Ustilago genus, has intensified diversified research focused microbe-based production,...
Speech based diagnosis-aid tools for depression typically depend on few and small datasets, that are expensive to collect. The limited availability of training data poses a limitation the quality these systems can achieve. An unexplored alternative large scale source vlogs collected from online multimedia repositories. Along with automation mining process, it is necessary automate labeling process too.In this work, we propose framework automatically label corpus in-the-wild possibly...
Speech is a complex bio-signal that has the potential to provide rich bio-marker for health. It enables development of non-invasive routes early diagnosis and monitoring speech affecting diseases, such as ones studied in this work: Depression, Parkinson's Disease. However, major limitation current based tools lack large diverse datasets. Existing datasets are small, collected under very controlled conditions. As such, there an upper bound complexity models can be trained using these There...
Polarity detection is a research topic of major interest, with many applications including detecting the polarity product reviews. However, in some cases, reviews might not be available while itself be, prohibiting use any form fully supervised learning technique. This scenario, different, close to that multiple instance (MIL). In this work we propose two new adaptations support vector machines (SVM) for MIL, θ-MIL, suit and infer products We perform experiments on proposed methods using...
Automatic detection of speech affecting (SA) diseases has received significant attention, particularly in clinical scenarios. However, the same task in-the-wild conditions is often neglected, part, due to lack appropriate datasets.In this work, we present in-the-Wild Speech Medical (WSM) Corpus, a collection videos, featuring subjects potentially affected by SA disease - specifically, depression or Parkinson's disease. The WSM Corpus contains total 928 and over 131 hours speech. Each video...