This dissertation advances generative speech enhancement by investigating both unsupervised and supervised machine learning approaches, with a focus on integrating visual information to improve robustness. The work is organized into three main contributions: The first contribution focuses on unsupervised generative speech enhancement. We explore a Bayesian framework combining variational autoencoders (VAEs) trained on clean speech…
This work provides a precise performance analysis of precoding in the asymptotic regime of the multi-user massive multiple-input multiple-output (MIMO) setting. Considering practical challenges, such precoders are typically formulated as non-linear, non-smooth optimization problems without closed-form solutions, often followed by non-linear post-processing operations. These aspects make it difficult to analyze the system behavior in multiple…
Probabilistic graphical models are a powerful concept for representing complex relations between components in systems with uncertain behavior. Stemming originally from statistical mechanics, their applications stretch across various fields such as computer vision, speech recognition, social network analysis, and many more. Often they allow for a compact formulation of practical challenges in terms of fundamental…
This thesis presents a corpus-based, comparative analysis of error patterns in human and automatic speech recognition (ASR based on utterances taken from spontaneous, unscripted face-to-face conversations. The utterances reflect specific patterns that are characteristic of this speaking style: They are disfluent through either a pause, a filler particle (FP), a break in the syntax, or…
Historical worship spaces, such as churches, cathedrals, mosques, and temples, possess unique acoustics that shaped the way rituals and music were developed and performed in these spaces. The acoustics of these spaces are often at risk due to renovations, structural decay, natural disasters, and changing usage patterns. This thesis explores how the acoustics of historical…
Biometric face authentication leverages the unique biological features of an individual?s face, providing a secure and convenient alternative to traditional password-based authentication. With the widespread adoption of face verification in remote authentication services and portable devices, ensuring the robustness of these systems against spoofing attacks has become increasingly critical. While traditional biometric threat models primarily…
The aim of immersive spatial audio technologies, as used, e.g., in virtual and augmented reality applications, is to provide the user with an immersive and plausible listening experience. The overall goal is to render the presented three-dimensional sound scenes realistically in a perceptual sense, either over headphones or using a multi-channel loudspeaker setup. Besides a…
Technology advancements profoundly impact numerous aspects of life, including how we communicate and interact. For instance, hearing aids enable hearing-impaired or elderly people to participate comfortably in daily conversations; telecommunications equipment lifts distance constraints, enabling people to communicate remotely; smart machines are developed to interact with humans by understanding and responding to their instructions. These…
Super-resolution (SR) is an ill-posed inverse problem focused on reconstructing high-resolution images from low-resolution counterparts by recovering missing details. Despite advancements, SR faces persistent challenges in generalization, balancing fidelity and perceptual quality, mitigating artifacts, and ensuring trustworthy results. This thesis tackles these issues through innovations in model architecture, loss design, and sample selection. Central to…
Goals and motivation: The PhD aimed at proposing new texture feature extraction algorithms based on information theory. More precisely, new entropy-based algorithms (EBA) were proposed to study the texture of biomedical datasets and, through the use of artificial intelligence techniques, to help diagnose pulmonary pathologies. Introduction: In 1948, Shannon introduced entropy to quantify a signal?s…
