A Neuromorphic Trigger for Efficient Audio Event Detection
For practitioners of real-time and resource-constrained audio processing, this work provides a low-cost front-end filter that significantly reduces computational load while maintaining high detection performance.
The paper introduces a neuromorphic trigger based on a spiking neural network that selectively gates input to downstream models for efficient audio event detection. On Anomalous Sound Detection, it achieves an F1 score of 0.97, and on Sound Event Detection, it enables a 42.6× reduction in FLOPs while improving the event-based error rate from 0.41 to 0.25.
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.