همایش بین المللی هوش مصنوعی

صفحه اصلی / دومین همایش بین المللی هوش مصنوعی

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

نویسندگان :

Hossein Sharify¹ Behnam Raoufi² Mohamad Mahdee Ramezanee³ Khosrow Hajsadeghi⁴ Saeed Bagheri Shouraki⁵

1- PhD from Sharif University of Technology 2- PhD from Sharif University of Technology 3- PhD from Sharif University of Technology 4- PhD from Sharif University of Technology 5- PhD from Sharif University of Technology

کلمات کلیدی :

acoustic scene classification،knowledge distillation،teacher ensemble،model compression

چکیده :

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.