Pattern Recognition Letters | Deep Learning for Acoustic Sensor Array Processing

DL-ASAP 2022

Artificial Intelligence



Dear Colleagues,
Acoustic sensor array processing is a well-studied field that has provided solutions to a wide range of practical problems such as source detection, estimation of source number, localization and tracking, source separation and signal enhancement, acoustic recognition, noise reduction and dereverberation. Although traditional multichannel signal processing methods reached a high level of maturity from a theoretical prospective and have shown to perform fairly well in simple applications, acoustic sensing in complex real-world applications is still a challenging problem. Reverberation, complex noise fields, dynamic reconfiguration of the acoustic scene, interferences, and concurrent multiple sources, represent today some of the most challenging problems in acoustic sensor array processing.
Recently, we have witnessed a growing interest in using artificial intelligence combined with sensor arrays to potentially solve acoustic sensing problems in complex environments and in emerging applications. Learning-based methods have shown to be able to exploit the multidimensional characteristics of a sensor array and marked the way to new solutions and novel applications.
The proposed special issue aims to present recent advances in the development of artificial intelligence and deep learning methods for acoustic sensor array processing emphasizing the associated theory, models, and applications. Automatic computer audition and microphone arrays need novel methods that use modern deep learning array processing addressing the challenges raised by real-life applications. The Special Issue welcomes research papers covering innovative learning-based approaches, theoretical advances, technological improvements, and novel applications in the field.
Guest Editors
Daniele Salvati, Managing Guest Editor
Maximo Cobos, Guest Editor
Fabio Antonacci, Guest Editor
Carlo Drioli, Guest Editor
Topics of interest
Machine learning models and algorithms for acoustic sensing
Detection and recognition of acoustic events
Deep learning for localization in noisy and reverberant environments
Data-driven tracking moving sources
Deep learning for acoustic scene analysis
Source separation and audio signal enhancement with expert systems
Voice recognition
Machine learning acoustic source identification
Binaural processing and artificial intelligence
Distributed acoustic sensor networks with expert systems
Deep learning for simultaneous localization and mapping (SLAM)
Source and scene classification
Important Dates
Submission deadline: March 20, 2022
Acceptance deadline: January 31, 2023