Event-driven spectrotemporal feature extraction and classification using a silicon cochlea model

Ying Xu; Samalika Perera; Yeshwanth Bethi; Saeed Afshar; André van Schaik

doi:10.3389/fnins.2023.1125210

Event-driven spectrotemporal feature extraction and classification using a silicon cochlea model

Front Neurosci. 2023 Apr 18:17:1125210. doi: 10.3389/fnins.2023.1125210. eCollection 2023.

Authors

Ying Xu¹, Samalika Perera¹, Yeshwanth Bethi¹, Saeed Afshar¹, André van Schaik¹

Affiliation

¹ International Centre for Neuromorphic Systems, The MARCS Institute for Brain, Behavior, and Development, Western Sydney University, Kingswood, NSW, Australia.

Abstract

This paper presents a reconfigurable digital implementation of an event-based binaural cochlear system on a Field Programmable Gate Array (FPGA). It consists of a pair of the Cascade of Asymmetric Resonators with Fast Acting Compression (CAR-FAC) cochlea models and leaky integrate-and-fire (LIF) neurons. Additionally, we propose an event-driven SpectroTemporal Receptive Field (STRF) Feature Extraction using Adaptive Selection Thresholds (FEAST). It is tested on the TIDIGTIS benchmark and compared with current event-based auditory signal processing approaches and neural networks.

Keywords: CAR-FAC; FEAST; LIF; STRF; electronic cochlea; event-based feature extraction; neuromorphic engineering.