Editorial: Bio-inspired Audio Processing, Models and Systems

Submitted by mounya on Wed, 01/15/2020 - 09:44
Authors
Shih-Chii Liu
Publication journal
Frontiers in Neuroscience

Bio-inspired systems look at biology to inspire engineering solutions that help explain, emulate and complement the intricate processes that take place in a biological system. As such, they operate at the intersection of biology and engineering and leverage advantages from both disciplines. When applied to brain sciences, bio-inspired systems often use non-conventional approaches to solve complex sensory and cognitive tasks. Recent developments in sensor design, algorithmic configurations, and network-level processing show the promise and efficacy of brain-like systems in solving complex tasks. While vision systems are widely explored in neuromorphic engineering design, audio systems offer unique challenges. These include careful handling of the time and space dimensions, issues related to temporal sampling and signal representation in both time and frequency, leveraging the redundancy in audio signals for complex detection and recognition tasks, as well as robust processing against noise and other interferers and maskers. Our auditory systems have evolved highly efficient solutions to audio scene analysis, spatial understanding, and sound recognition. We wish to better understand the biological solutions that allow the brain to process sounds in unknown and highly distorted conditions; in order to help advance state-of-art audio systems that often operate well under well-controlled environments but fail to generalize, adapt and efficiently process unknown conditions. Furthermore, we want to apply engineering methods to better understand biological processes, using non-invasive methods. By leveraging both our knowledge of the biology in building better systems, as well as new technological advantages to unravel secrets of the brain, we hope to enrich the conversation across both disciplines in order to advance our understanding of the brain function and help improve technologies that impact our lives in a wide range of domains.

Keywords
neuronal networks