A Low-Complexity Deep Learning Framework For Acoustic Scene Classification

Abstract

In this technical report, a low-complexity deep learning system for acoustic scene classification (ASC) is presented. The proposed system comprises two main phases: (Phase I) Training a teacher network; and (Phase II) training a student network using distilled knowledge from the teacher. In the first phase, the teacher, which presents a large footprint model, is trained. After training the teacher, the embeddings, which are the feature map of the second last layer of the teacher, are extracted. In the second phase, the student network, which presents a low complexity model, is trained with the embeddings extracted from the teacher. Our experiments conducted on DCASE 2023 Task 1 Development dataset have fulfilled the requirement of low-complexity and achieved the best classification accuracy of 57.4%, improving DCASE baseline by 14.5%.

Robust, General, and Low Complexity Acoustic Scene Classification Systems and An Effective Visualization for Presenting a Sound Scene Context

“…All three models are evaluated on a wide range of ASC datasets mentioned in Section II: DCASE 2018 Task 1A, DCASE 2018 Task 1B, DCASE 2019 Task 1A, DCASE 2019 Task 1B, DCASE 2020 Task 1A. For DCASE 2021 Task 1A and DCASE 2020 Task 1, we report the results which are from our submitted models presented in [100] and [101], respectively. Notably, the submitted models also make use of CNN-based network architecture and model compression techniques of channel reduction (CR), channel deconvolution (CD), and Quantization (Qu.…”