Explore the emerging field of computational analysis of sound scenes in this comprehensive lecture by Professor Tuomas Virtanen from Tampere University. Gain an overview of recent developments, including basic scene classification, event detection, and advanced tasks like joint localization and audio captioning. Discover state-of-the-art machine learning techniques based on deep neural networks used to solve these tasks, and learn about the data required for training such systems. Examine examples of machine learning architectures and findings from the latest DCASE evaluation campaigns. Delve into open research questions in this rapidly evolving field, presented by a pioneering expert in single-channel sound source separation and environmental sound analysis.