A Neuromorphic Trigger for Efficient Audio Event Detection

2026-06-16

AI Summary

This work addresses the challenge of efficiently processing continuous audio streams in real-time and resource-constrained systems by proposing a lightweight neuromorphic trigger based on a fully connected spiking neural network (SNN). Acting as an efficient front-end, the proposed trigger selectively forwards only salient audio segments to downstream, computationally expensive models. To the best of our knowledge, this is the first application of a neuromorphic trigger to audio event detection, achieving high-precision, class-agnostic selection of relevant audio regions. Experimental results on the URBAN-SED and DCASE 2017 datasets demonstrate that the system maintains a high F1 score of 0.97 while reducing computational cost by a factor of 42.6 in FLOPs and lowering the event-level error rate bound from 0.41 to 0.25, substantially decreasing overall computational overhead.

Abstract

Efficient processing of continuous audio streams remains a key challenge for real-time and resource-constrained systems. This paper introduces a neuromorphic trigger for audio event detection, based on a spiking neural network (SNN) that selectively gates input to downstream models. The proposed trigger acts as a low-cost front-end, identifying salient audio segments and forwarding only these to a more computationally intensive model for tasks such as classification. The trigger is implemented as a lightweight fully connected SNN and evaluated on two representative tasks: Anomalous Sound Detection (ASD) and Sound Event Detection (SED). For ASD, the trigger achieves a one-second segment-based F1 score of 0.97 on a class-agnostic form of the URBAN-SED dataset, demonstrating high reliability in identifying relevant audio regions. For SED, the trigger is combined with the Dang classifier on the DCASE 2017 Challenge Task 2 dataset, showing a potential $42.6\times$ reduction in FLOPs while reducing the lower bound of the event-based error rate from 0.41 to 0.25. These results highlight the potential of neuromorphic triggers as real-time, energy-efficient front-end filters, enabling substantial reductions in computational cost.

Problem

• audio event detection
• real-time processing
• resource-constrained systems
• efficient audio processing

Innovation

• neuromorphic trigger
• spiking neural network
• audio event detection
• computational efficiency
• low-power front-end