Ensemble-Guided Distillation for Compact and Robust Acoustic Scene Classification on Edge Devices

TL;DR

This paper introduces a compact, robust acoustic scene classification framework using ensemble-guided knowledge distillation, optimized for edge devices, achieving state-of-the-art results on a benchmark dataset.

Contribution

It proposes a novel ensemble-guided distillation method with a lightweight student network and diverse teacher ensemble for efficient edge deployment.

Findings

Achieves state-of-the-art accuracy on TAU Urban Acoustic Scenes 2022 Mobile benchmark.

Demonstrates robustness to device and noise variability.

Enables efficient inference suitable for edge devices.

Abstract

We present a compact, quantization-ready acoustic scene classification (ASC) framework that couples an efficient student network with a learned teacher ensemble and knowledge distillation. The student backbone uses stacked depthwise-separable "expand-depthwise-project" blocks with global response normalization to stabilize training and improve robustness to device and noise variability, while a global pooling head yields class logits for efficient edge inference. To inject richer inductive bias, we assemble a diverse set of teacher models and learn two complementary fusion heads: z1, which predicts per-teacher mixture weights using a student-style backbone, and z2, a lightweight MLP that performs per-class logit fusion. The student is distilled from the ensemble via temperature-scaled soft targets combined with hard labels, enabling it to approximate the ensemble's decision geometry with a single compact model. Evaluated on the TAU Urban Acoustic Scenes 2022 Mobile benchmark, our approach achieves state-of-the-art (SOTA) results on the TAU dataset under matched edge-deployment constraints, demonstrating strong performance and practicality for mobile ASC.