معرفی
Annamaria Mesaros is an Associate Professor (tenure track) and Academy Research Fellow in Computing Sciences. Her research focuses on sound event detection, acoustic scene classification, and binaural audio processing, with significant contributions to reproducibility in technical reporting.
- Education: Doctor of Science (Technology) in Information Technology (2012)
Research Interests: She specializes in machine learning applications for audio signal processing, including class-incremental learning, distance estimation, and multilingual audio captioning. Her work addresses challenges in device-specific domain adaptation and accessibility for vision-impaired users.
Scientific Contributions: Mesaros has published extensively on IEEE/ACM Transactions and co-created benchmark datasets like TUT Urban Acoustic Scenes (2018) and TAU Urban Acoustic Scenes (2019–2024). Her recent 2024–2025 publications highlight advancements in sound source localization and incremental learning frameworks.
Collaborations: She frequently collaborates with researchers like Heittola, Virtanen, and Politis, and has organized key DCASE workshops (e.g., 2019, 2023).
