Paul Calamia: Reality Labs Research (Meta), USA
Enzo De Sena: University of Surrey, UK
Wenyu Jin: Cruise LLC, USA
Toon van Waterschoot: KU Leuven, Belgium
Submission Status: Open | Submission Deadline: 31 January 2024
EURASIP Journal on Audio, Speech, and Music Processing is calling for submissions to our Collection on Data-driven and Machine Learning-based Room Acoustic Estimation and Modeling.
Reverberation is a near-constant presence in our daily lives, and being able to estimate, synthesize, control or suppress reverberation is therefore important. The field of room-acoustic modeling is rapidly adopting data-driven methods, including machine learning (ML), due to technical advancements in related fields and the availability of increasingly larger amounts of computing power and data. Given enough data, mathematical models can be trained to discover underlying acoustic representations useful to perform the task at hand. However, compared to more traditional signal-processing methods, ML and data-driven methods also bring a multitude of unique challenges to consider to increase their utility, generalisability, reliability, and interpretability.