معرفی
Alessandro Rossi is a researcher at Fondazione Bruno Kessler (FBK) in Trento, Italy, affiliated with the SpeechTek group. His work focuses on developing deep learning approaches for speaker diarization and acoustic modeling. Previously, he collaborated with the UDOO Team at the University of Siena's Mind Lab, investigating machine learning solutions for novelty detection in audio/video surveillance.
He holds a Ph.D. in Machine Learning (2017) from the University of Siena's Department of Information Engineering and Mathematical Sciences, supervised by Prof. Marco Gori and Prof. Marco Maggini. His doctoral research explored regularization methods, temporal coherence constraints, and manifold regularization in semi-supervised learning for video understanding tasks.
Research Focus
Rossi's expertise spans:
- Core Machine Learning: Regularization methods, temporal dynamics, semi-supervised learning
- Deep Learning Architectures: Applied to acoustic modeling and speaker recognition
- Multimedia Analysis: Novelty detection in surveillance systems
Affiliations
He is currently active in the SpeechTek research group at FBK. Past affiliations include the UDOO Team at the University of Siena's Mind Lab (DISPOC department).
