An Exciting Juncture: The Convergence of Machine Learning and Signal Processing

This is our sixth and final issue of 2024. It is hard to believe that a year has gone by since our term as the new editorial team started in January. In our first year, in addition to our usual array of technical overviews and Society news, we addressed a number of topics of significance for our community in the hopes of starting a discussion. A key theme among those has been the relationship of machine learning (ML)—or more broadly, artificial intelligence (AI)—to signal processing (SP) research and education. Now, we conclude 2024 with a special issue dedicated to articles that propose solutions for audio SP at the intersection of model-based and data-driven solutions. Like many other active areas of research these days, in audio SP as well, methods based on ML are the prominent solutions. However, the collection of articles our guest editors have carefully selected all emphasize the importance of achieving a balance between the data-driven and model-based approaches [A1], [A2], [A3], [A4], [A5], [A6], [A7]. These articles demonstrate the importance of not solely relying on learning from the data but taking the physical aspects of the problem into account to enable efficient and effective solutions that are also explainable and interpretable.